Skip to main content

Full text of "The Races Of Europe"

See other formats

64022 > 


n.ciui.'Tronder.fitc.. un- 
/, rzed 

:ed, brachycephalic 


>r (TeviecType.ek) 


A A A " A A A 

A A A A A A A 
A A A A A A A 
A A y^ A A A A 
A / /* A^ 

A A ^ A 

A A A A A 

A A A A A A 

A A A A 

. ., AAAAA 



A \ A A A A A A A A A 
A A A A A A 
A A A 
A A 

O O O O Oi O O O O O O O O O 


o o 01 o o o OOGOOQO *> $. 







The Races 
of Europe 


Assistant Professor of Anthropology 


New Tork 






Published, April, 1939. 



Author of the earlier and classic 


at whose suggestion the present work was begun 
and in whose honor it is named. 


The present book is offered to the College audience as a text in a specific 
branch of physical anthropology. In it an attempt is made to trace the 
racial history of the white division of Homo sapiens from its Pleistocene 
beginnings to the present. Although six chapters are specifically devoted 
to a study of skeletal material by consecutive cultural periods, the main 
emphasis is placed upon the racial identification and classification of 
living white peoples. If there is one consistent theme in this book, it is 
that physical anthropology cannot be divorced from cultural and his- 
torical associations, and that there is no such thing as "pure" biology, 
at least in reference to human beings. 

In writing a book of this character it has been necessary to employ a 
number of technical terms; the reader will find these defined in the glos- 
sary. Statistical tables have been purposely omitted from the text, but 
since many of the conclusions and identifications made in the chapters 
of skeletal history are novel, it has seemed advisable to document them 
by means of tabular material. For this reason the fifty-three columns of 
basic cranial means have been included as Appendix I. 

References to all sources from which material, anthropometric or other- 
wise, has been drawn are given in footnotes in the sections in which 
specific data are mentioned. Although over four thousand titles have been 
consulted in the preparation of this volume, the author makes no pre- 
tense to have covered the entire literature of the subject. A number of 
unimportant references have been purposely omitted, and many others 
which are important have without doubt been overlooked. Except for 
materials used with special permission in advance of publication, no 
reference is made to data appearing later than July, 1938. 

Two collateral phases of physical anthropology have, for adequate 
reasons, been completely avoided: the study of blood groups and the 
question of racial intelligence or racial psychology. The science of blood 
groups has, by 1 938, developed a prodigious bibliography of its own, and 
will soon be treated in a special survey by Professor Wm. Boyd of Boston 
University. So far as specialists in this field have yet determined, there is 
no genetic linkage between blood group types and anthropometric 
phenomena. The subject of racial intelligence has, on the other hand, 
not progressed far enough to merit inclusion in a general work of racial 
history; it has furthermore provided too ready a field for political exploita- 
tion to be treated or interpreted as a side issue with scientific detachment. 


Races, in the present volume, are studied without implication of inferiority 
or superiority. 

In the financing of the work, in the collection of data, and in the prepa- 
ration of the manuscript, many persons have participated. The initial 
work of collection and preparation was financed, for two years, by gener- 
ous grants from the Milton Fund and John G. Clark Bequest of Harvard 
University. Further financing which permitted its completion was pro- 
vided by my father, Mr. John Lewis Coon, by The Macmillan Company, 
and by Mr. Lloyd Cabot Briggs. For the original suggestion that I be 
chosen to write the book, for his support in obtaining the original research 
funds, and for his continual advice and encouragement, I am deeply 
indebted to my teacher, Professor Earnest A. Hooton, who initiated me 
to physical anthropology and to whom I wish to render here an expression 
of homage and appreciation not only as my personal mentor but also as 
the spiritual father of American physical anthropology. 

Of the many assistants who helped with the tedious labor of translating, 
abstracting, calculating, plotting, checking, and typing, four deserve 
especial credit: Mrs. Mary Ruby Gardner, Miss Anna Szugzda, Mr. 
Eugene C. Worman, and Mr. Jens Yde. Mr. Elmer Rising, who pre- 
pared all of the maps, charts, and line drawings, made the task of illus- 
tration easy with his experience and cooperation. Mr. Frederick P. 
Orchard and Miss Marion Lambert assisted in the preparation of the 
photographic illustrations. 

Miss Constance Ashenden, Librarian of the Peabody Museum of 
Harvard University, under whose direction every article in the scientific 
periodicals included in the library has been separately catalogued by 
author, subject, and country, placed at my disposal her great knowledge 
of the bibliography of anthropology, as well as her time and patience. 
To her and to Mr. Francis Gould, her assistant, I owe an especial debt 
of gratitude. 

The following persons have permitted me to make use of unpublished 
anthropometric materials: Dr. Gordon T. Bowles, Mr. C. Wesley Duper- 
tuis, Mr. Robert W. Ehrich, Dr. Henry Field, Mr. James Gaul, Mr. 
Herbert R. Glodt, Dr. Earnest A. Hooton, Dr. Byron O. Hughes, Dr. 
Frederick P. Hulse, Dr. W. Marion Krogman, Mr. Homer H. Kidder, 
Mr. Martin Luther, Dr. Theodore W. McCown, Dr. Geoffrey M. Morant, 
Dr. Carl C. Seltzer, Dr. William Shanklin, Professor Boris N. Vishnevsky, 
Mrs. Ruth Sawtelle Wallis, Professor Franz Weidenreich. Each of these 
persons will be further accredited in reference to the specific material used. 
It is hoped that a cursory mention of their data in this volume will stim- 
ulate interest in their detailed monographs which will follow. Needless 
to say, none of them is to be held responsible for any erroneous or un- 


warranted interpretations which I may have placed on their materials. 

I wish also to thank in this place those persons and institutions which 
have permitted me to reproduce photographs and paintings. Individual 
Credit will be given in each instance. The majority of the photographs 
used in this book, however, were taken by the author, with the generous 
assistance of many people. These include especially Miss Marion Black- 
well, director of the International Institute in Boston, and her assistant 
Miss Olga St. Ivanyi; Mr. Arthur Megerdichian; Mr. Phillip Way and 
Mr. Merico Petrolati, of the Ludlow Manufacturing Company, Ludlow, 
Mass. ; Mr. Bror Tamm, Mr. H. W. Johnson, and the owners of the ship- 
building firm of George Lawley and Son; Mr. Ian Drysdale of the A. C, 
Lawrence Leather Company of Peabody, Mass.; Mr. Michel Abourjaily, 
of Boston; M. Dumas, of the Dumas Bookshop, Boston; Mr. Heinrich 
Wolff, manager of Gundlachs Hofbrauhaus in Boston; Father Jan Kozit- 
sky; Mr. John Brunswick and the officers of the Czechoslovakian Club 
of Boston; Mr. James Stragunas; and numerous others, including all 
whose photographs appear in the plates illustrating racial types. 

For specific advice and assistance, I have especial reason to be grateful 
to the following: Professor Glover Allen, for advice concerning fauna; 
Dr. Gordon T. Bowles, for the preparation of Map 16, and for information 
concerning the peoples of Iran, Afghanistan, and India; Professor Kirk 
Bryan, for information concerning Pleistocene and post-Pleistocene cli- 
mate; Professor V. Gordon Childe, for reading the manuscript of Chapters 

II through VII, and for suggesting many important changes; Dr. Vladi- 
mir J. Fewkes, for preparing Maps 2 and 3, and for much advice upon 
the European archaeology of the Neolithic, Bronze, and Iron Ages, and 
for data and advice on the subject of Slavic history; Dr. H. O'Neill 
Mencken, for advice concerning the archaeology of the Iron Age, and of 
the British Isles in particular; to Mr. Gabriel Lasker, for aid in preparing 
the glossary; Dr. J. R. de la H. Marett, for ideas and stimulation on 
the subject of human evolution; Professor William M. McGovern, for 
permitting me to read the manuscript of his "Early Empires of Central 
Asia," and for advice on the subject of Central Asiatic history; Dr. Hal- 
lam L. Movius, for assistance in the preparation of Map 1 and Figure 16, 
as well as in the writing of Chapters II and III; Dr. Robert W. Pfeiffer, 
for data on early Jewish history; Professor J. Dyneley Prince, for expert 
opinion on the question of Sumerian linguistics; Professor George Sarton, 
for advice on the handling of references; Mr. Vilhjalmur Stefansson and 
Mr. Charles Harding, for advice and data on the subject of the Norse- 
men; Mr. Lauriston Ward, Mr. James Gaul, and Mr. D. W. Lockard, 
for supervision and assistance on the subject of Near Eastern archaeology; 
Professor Harry Wolfson, for an elucidation of Jewish history and assist- 


ance in preparing the sections on the Jews. To this list must be added the 
names of Professor M. F. Ashley- Montagu, Professor W. M. Krogman, 
and Dr. H. L. Shapiro, who read the book in galley proof and are re- 
sponsible for many necessary changes, deletions, and additions. 

As the reader will readily perceive, the experts listed above, most of 
whom are already renowned as illustrious scholars, have had no small 
part to play in the preparation of this volume. To them singly and 
collectively I owe a debt which it will be impossible for me to repay, 
and to them I offer my apologies if I have betrayed their generosity and 
their competence. 

To Their Majesties the Kings of Yemen and Albania, and to His 
Highness the Sultan of Mukalla, I also wish to express my gratitude for 
permission and assistance in the collection of data which are here pre- 
sented for the first time. 

Finally, to the officers and staff of The Macmillan Company, I am deeply 
indebted for their generosity, cooperation, and forbearance. 

C. S. G. 

February, 1939. 








(1) INTRODUCING Homo Sapiens 16-18 






WHOLE 28-33 
























(3) IRAN AND IRAQ 86-91 





NORTH 101-104 







(14) CONCLUSIONS 126-130 


(1) INTRODUCTION 131-135 


(3) THE MINOANS 140-142 

(4) THE GREEKS 142-146 





BEAKERS 154-157 









(2) THE. ILLYRIANS 182-186 

(3) THE KELTS 186-193 

(4) THE ROMANS 193-195 

(5) THE SCYTHIANS 195-201 

(6) THE GERMANIC PEOPLES ... 201-216 

(7) THE SLAVS 216-220 

(8) CONCLUSIONS 220-222 



(Speakers of Uralic and Altaic) PAOES 

(1) THE FINNO-UGRIANS 223-226 


















(1) INTRODUCTION 297-298 

(2) THE LAPPS 298-306 

(3) THE SAMOYEDS 306-307 


(5) ICELAND 323-326 

(6) SWEDEN 326-332 

(7) DENMARK 332-337 



(10) THE BALTIC FINNS: Livs AND ESTHS 351-355 

(11) THE BALTIC FINNS: FINLAND . ". 355-359 


(13) CONCLUSIONS 368-369 



(2) IRELAND 376-384 






(1) INTRODUCTION 400-401 











(12) THE TUAREG 471-474 





(17) THE BASQUES 501-504 

(18) THE GYPSIES 504-507 

(19) CONCLUSIONS 507-509 


(1) INTRODUCTION 510-511 

(2) FRANCE 511-522 

(3) BELGIUM 522-529 


(5) GERMANY 535-547 


(7) ITALY 554-559 


(a) CZECHS AND WENDS 559-563 


(b) POLAND AND RUSSIA 563-576 


(11) THE MAGYARS 584-586 




(14) THE GREEKS 604-609 


(15) BULGARIA 609-612 


(17) THE OSMANLI TURKS 617-622 





(21) CONCLUSIONS 646-648 





TERS II-VII 655-665 

II. GLOSSARY .... . . . .... 666-683 


IV. LIST OF BOOKS 692-700 

INDEXES' 701-739 







5. STATURE 252-253 

6. CEPHALIC INDEX 258-259 

7. HEAD SIZE 262-263 











Chapter I 



The present book is a textbook designed for the use of college students 
who have had or are taking a preliminary course in anthropology. Enough 
of it is, however, written in a non-technical way, so that students of allied 
disciplines may use it for reference. The subject matter to be studied con- 
sists of the body of statistical material collected by the world's physical an- 
thropologists which concern the somatic character of peoples belonging to 
the white race. This material may be divided into (A), skeletons; and (B), 
metrical data and observations on the living. 

By the use of this material we propose to follow the history of the white 
race from its Pleistocene 1 beginnings to the present, and to provide a classi- 
fication of sub-races which will be fully in accord with the facts as we now 
know them. We submit the thesis that man, as a domestic animal, is 
extremely variable; and that he has subjected himself, in his wanderings, 
to all of the environments of the earth, and hence is subject to environ- 
mental modification in a way unequalled by any other species. We further 
suggest that man, through his development of human cultures, has modi- 
fied his bodily form by his own devices. 

During the Pleistocene period there were several species of primates 
which had attained some degree of human culture, by the acquisition of 
stone implements, of fire, and of speech. In the present post-glacial or 
interglacial period, in conformity with the general reduction in faunal 
varieties, man has been reduced to a single species, unique in a single 
genus. During the Pleistocene one species, at least, had developed in the 
manner of a foetalized terrestrial ape, and it is that species which carries 
today the main stem of Homo sapiens. Other species, including the fossil 
men of Java, of Peking, and Homo wander thalensis, had developed at the 
same time into a heavier, hypermasculine endocrine form, with a luxuri- 
ance of jaws, teeth, and bony crests. 

We propose to demonstrate that these non-foetalized species did not 
wholly die out, but that at least one of them was absorbed into the main 

1 The term Pleistocene is used here to signify the time span which, in Europe, began 
with the advance of the first Quatenary glaciation and which ended with the retreat of 
Wurai II. 



human stem, at some time during the Middle, or the initial part of the Late, 
Pleistocene. From this amalgamation was produced the large, rugged, 
and relatively un-foetalized group of Upper Palaeolithic men in Europe, 
North Africa, and northern Asia. This type of man passed over Bering 
Straits in early post-glacial times, if not earlier, to provide the basic ge- 
netic stock from which the American Indian developed, in combination 
with later arrivals. From a branch of this hyperborean group there evolved, 
in northern Asia, the ancestral strain of the entire specialized mongoloid 

We suggest that the ancestors of the whites in their major form developed 
during pluvial periods of the Pleistocene in parts of what is now the arid 
zone reaching from the Sahara to northern India; that in post-glacial times 
many were forced out of these homes by desiccation, and that some of 
them originated agriculture and animal husbandry in northeastern Africa 
and southwestern Asia. From these centers agricultural pioneers followed 
post-glacial zones of climate into Europe, gradually encroaching upon 
the lands formerly glaciated. In most of the regions which they occupied 
they greatly outnumbered the descendants of the hunters and fishers 
whose ancestors had clung on since glacial times, and many of whom had 
followed the retreating ice toward its last melting nuclei. 

The occupation of all arable lands, and those suitable for grazing, was 
not completed in a century, or in a millennium; the process was a gradual 
one, and the withdrawal of the earlier inhabitants into environmentally 
protected fastnesses equally gradual. The entry of food-producers from 
Asia and Africa did not take a single route or involve a single people; it 
was a complex sequence of migrations through several ports of entry. 
The various strains of food-producers mixed with the food-gatherers 
whom they encountered, and with each other, until, in our own time, 
not a single group of complete food-gatherers has remained in white man's 

The food-producers seem to have been variants on one central racial 
theme, the basic Mediterranean. This basic Mediterranean stock varied 
in many respects, especially in stature and in pigmentation, but in its 
essential qualities, which segregated it from non- whites, it was remarkably 
uniform. We do not know that the survivors of the food-gatherers whom 
the Mediterranean food-producers absorbed were white in soft-part 
morphology, and there is some evidence that some had begun to evolve 
in a mongoloid, others perhaps in a negroid, direction. Such variations 
may be seen within the present composite white racial amalgam. 

At any rate, the main conclusion of this study will be that the present 
races of Europe are derived from a blend of (A) , food-producing peoples from Asia 
and Africa, of basically Mediterranean racial form, with (j5), the descendants of 


inter glacial and glacial food-gatherers, produced in turn by a blending of basic 
Homo sapiens, related to the remote ancestor of the Mediterraneans, with some o- 
sapiens species of general Neanderthaloid form. The actions and interactions of 
environment, selection, migration, and human culture upon the various entities within 
this amalgam, have produced the white race in its present complexity. 

In view of these circumstances, the exact classification of living whites 
into sub-races, such as Nordics, Alpines, Binaries, and so on, need not be 
made at this point, but can await (A) the historical study of the white 
race which will follow in Chapters II to VII; and (B) the survey of the 
living as a whole which will be made in Chapter VIII. In Chapters IX 
to XII, inclusive, we will make a more detailed regional survey of the 
living peoples of Europe to supplement the preceding sections. 


Before proceeding to a detailed historical survey or to technical mat- 
ters, it seems advisable to state at greater length than in the preceding 
section some of the principles which we believe to govern the formation of 
human races. First of all the question arises, "What is a race?" and the 
problem of this definition must be squarely faced. In the course of the 
present study the author has developed a definite point of view on this 
subject, which may be expressed as follows: The concept race is a general 
one, and any attempt to chain it down to a more specific meaning repre- 
sents a too rigid attempt at taxonomy. The use, under strict definition, 
of such convenient words as sub-race, stock, variety, local type, etc., im- 
plies a Linnaean classification of categories which is foreign to the facts 
of human biological differentiation. 

One may, in a group of animals such as man, definitely name and 
classify the major group to which all individuals belong. All living va- 
rieties of human beings are mutually fertile, and there is no other animal 
with which man may be crossed. Although the fertility test is not neces- 
sarily a diagnostic, Homo sapiens in the living sense comprises, without ques- 
tion, a species, even if in the formation of the living human group more 
than one related species, now extinct in the pure form, was absorbed. 2 

So much for the larger group. Within this larger group there are many 
variations of superficially great importance. There are pygmy men whose 
mean stature is less than 150 cm. There are giant-like men whose mean 
stature is over 180 cm. At the same time there are black men and white 
men; men with kinky hair, men with straight hair; men with beards and 
without; and so on. Their variation is much greater than that found 
among wolves, or among tigers, or among any one species of mice. Yet 
it is not as great as the variation found among dogs, who again form a 

2 See Chapter II, section 5. 


single species, and who in turn may include a blend of two wolf and 

Here again, we must repeat, man is a domestic animal, and as such is 
subject to the laws which govern animals in domestication. Being less de- 
pendent in a direct sense upon a given environment than a wild animal, 
he is much more variable; having become numerous as a result of this 
partial emancipation, he has spread into many environments, so that 
what influences these environments have had upon him have been ex- 
tremely varied. At the same time the laws which govern his mating are 
different from those which govern the conjunction of wild animals. Fur- 
thermore there has been some degree of selection in this mating, but less 
than the selection which has so profoundly differentiated the dog. 

All of the principles mentioned above have produced, as their effect, a 
prodigious differentiation within the human species, and one which must 
at times have proceeded with startling rapidity. At the same time there 
has taken place an almost equally great mixing and blending of peoples, 
under circumstances that could hardly occur among wild animals. For 
example, the mixture between whites and negroes has most frequently in- 
volved white men and negro women, and only occasionally the reverse. 
Within the ranks of mixture, there has often been a selection on the basis 
of differential social values attached to different combinations of charac- 
ters. As a result of all these factors, one must not suppose that a racial 
classification of man into a simple and orderly scheme can be easy. 

We have already recognized the concept species in regard to man. There 
is one other concept, wholly theoretical for practical reasons, which may 
be recognized with equal definition. That is the pure strain, the result of 
generations of inbreeding and selection of recessive characters. In man, 
the pure strain is impossible to create unless our social system radically 
changes. In rats, guinea pigs, and fruit flies, it has been created. From 
rats, guinea pigs, and fruit flies, biologists slowly and painstakingly dis- 
cover the laws which govern inheritance. They almost unanimously favor 
the Mendelian form, and there can be little doubt that Mendelism also 
applies to man. But man is a genetically complex animal, and we do not, 
apparently, measure characters which are Mendelian units. If we were 
to measure the right things, we would theoretically find that Mendel's Law 
is always applicable. The principle of inheritance through blending, by 

A + B 

which is derived the formula , depends upon a multiplicity of com- 
pensating Mendelian characters. That these are not always multiple, or 
that they do not always compensate, is shown by certain instances in 
which blending has not resulted from mixture. 

For example, the height of the cranial vault and the heights of the face 


and nose often fail to respond in the expected manner. Negro-white 
hybrids in the United States have long faces and noses, 8 and so do Ethio- 
pians. 4 Pitcairn Islanders have more convex noses than do either English 
or Tahitians. 5 Other instances have been found in which human inherit- 
ance has failed to assume the character of a blend. These serve merely 
as examples. Mixture alone, however, cannot create and perpetuate a 
new racial form, although it can produce new combinations. Mixture 
when combined with selection, to emphasize the new and eliminate the 
old, can, however, produce a decisive change. 6 

In view of the complexity of the human species, as a result of its cultural 
peculiarities which have separated it from the rest of the animal world, it 
is not easy to define the word "race." Since man is the oldest domestic 
animal, his variation and selection have operated over an immensely 
longer span of time than those of the other species for whose present forms 
he is responsible. Any attempt to classify him by a rigid scheme is im- 
mensely difficult, and the scheme must be elastic if it is to work at all. 
Hence the term "race" must also be elastic. We may recognize, if we 
like, certain major races of the Old World such as the Khoi-San (Bush- 
man-Hottentot), the Pygmy, the Australoid, the Negro, the Mongoloid, 
and the White. Within each of these major racial groups there are, or 
have been, smaller entities which may deserve the designation of race in a 
lesser sense. These smaller entities consist, for the most part, of groups 
of people reasonably isolated, and developing into local physical enclaves 
by the three processes, usually linked, of amalgamation, selection, and en- 
vironmental (in the total sense, including cultural) response. At what border- 
line point such an entity becomes a major race, it is not always possible 
to say. 

Let us consider these three forces amalgamation, selection, and environ- 
mental response. We have already mentioned the first, which is more com- 
monly called race mixture. We have already observed that while blending 
seems to be the usual result, in some criteria there is evidence of simple 
Mendelism or the heaping of dominants or recessives. Amalgamation, 
furthermore, can produce a differential dominance based on age grading; 
for example, the dominance of hair blondism in infancy, coupled with the 
darkening of the hair in adolescence and adult life, link blondism with 
infantile characters. The same is not true of eye blondism, which grows 
slightly more pronounced with age. At the same time mongoloid morpho- 
logical characters are more pronounced in infantile hybrids than in the 

8 Hooton, E. A., HAS, vol. X, part II, 1932, pp. 42-107. 
4 Unpublished data in author's possession. 
6 Shapiro, H. L., The Heritage of the Bounty, pp. 229-233. 
6 Baur, Fischer, and Lenz, Human Heredity, p. 176. 


adults; the reverse is true of most distinctively white features in combina- 
tion with those of either negroids or mongoloids. This differential age 
dominance is, except in the case of blondism, an heritable endocrine 
function connected with the relative degree of infantilism associated with 
each of the major racial groups. 

Selection is a difficult force to study in man, at least in a scientific sense. 
But it is without question one of great importance. Sexual selection prob- 
ably has and always has had a certain application, which may be seen in 
the current standards of beauty in different countries. The standards of 
one group may be shifted through the cultural medium to another. But 
since in any population other than an industrial, civilized one there are few 
bachelors and few spinsters, sexual selection must have worked slowly in 
most cases, at least in the sense of an eliminative rather than a segregative 
principle. Warfare, again, kills off a selected group of males, while celibacy 
connected with the assumption of religious offices may render genetically 
ineffective a selected population element. 

The most important selection is probably that consequent on changes 
of environment, by which the selective factor may perhaps be a physio- 
logical economy in response to new types of mineral deficiency. This type 
of selection may have been of profound importance in the evolution of 
man as a species, as well as of different races. 7 Small, foetalized, relatively 
weak races may be more efficient and hence more suitable for survival in 
certain environments than larger, more muscular, and less infantile ones. 
Small, foetalized, and relatively defenseless mammals develop elaborate 
social devices by which the solidarity of the group compensates for the 
deficiency in individual aggressiveness; man on the whole is a social animal 
comparable in this respect to the Cebus monkey. The type of environ- 
mental selection postulated by Marett may have been of profound im- 
portance in the evolution of man as a species, as well as of different races. 

Another form of selection is intimately concerned with the complexity 
of the social structure. When a population is stratified into social horizons, 
this cultural differentiation is often the result of the conjunction of two or 
more social and hence ethnic groups, from two or more geographical 
sources. It takes time for cultures to blend and for people who practice 
these cultures to mix, and if there exists, at the same time, the idea that 
one group is superordinate and the other subordinate in social values, the 
social mechanism will often function in such a way as to perpetuate this 
cleavage. Thus the mixing process will be retarded, and at the same time 
a difference in the reproductive rates of the two Facially identified social 
horizons may arise. 

As a rule, at least in modern times, the group which is considered sub- 

7 Marett, J. R. de la H., Race, Sex, and Environment. 


ordinate will reproduce with greater fecundity than will the superior class. 
In this way the upper class will gradually disappear, or else social mobility 
will gradually replace the upper from the ranks of the lower, and the social 
distinction will remain, but without racial significance. Thus a differential 
reproductive rate has, in effect, a selective value, and one population may 
quietly replace another. Whether or not the replacement is complete, the 
relative numerical importance of the two genetic strains will have been 

Extreme differences in skin color, in body odor, and in face form are 
more active deterrents to such mobility than are differences important to 
the anthropologist but not to the public, such as the cephalic index and 
other measures of head form. Differences of the first class prevent the 
American Negro from complete absorption into the ranks of the white, for 
his diagnostic racial characters, unless the negroid factor in the individual 
inheritance is dilute, are easily noticeable. On the other hand differences 
in head form are not usually noticed, and a brachycephalic white popula- 
tion may replace a dolichocephalic one by means of social mobility. 

So far we have been considering selection within a geographically im- 
mobile group, or rather, selection at the geographical point under con- 
sideration. But there is still another type of selection which is very im- 
portant, and that is mobile selection, operating at the point of emigration, 
the source of population supply. We shall see, in our survey of prehistoric 
European racial movements, 8 that the Danubian agriculturalists of the 
Early Neolithic brought a food-producing economy into central Europe 
from the East. They perpetuated in the new European setting a physical 
type which was later supplanted in their original home. Several centuries 
later the Corded people, in the same way, came from southern Russia 
but there we first find them intermingled with other peoples, and the cul- 
tural factors which we think of as distinctively Corded are included in a 
larger cultural equipment. The Corded people, therefore, who left south- 
ern Russia and moved westward into central and northwestern Europe, 
were a selected group of people, choseii from a larger and more hetero- 
geneous human storehouse. This situation clearly involves the principle 
that people who migrate from an old home to a new do not represent, in most cases, 
the total or typical physical form of the home land, provided that the new home is 
different from the old; but they represent a special group selected on the basis of their 
suitability and opportunity for migrating. This principle can be clearly seen 
in the study of modern migrating peoples. 

The Poles who came to the United States during the nineteenth cen- 
tury, and the early decades of the twentieth, did not represent a cross- 
section of the Polish population, 9 but a taller, blonder, longer-headed 

s Chapter IV. * Rosinski, B., PAn, voL 8, 1934, pp. 42-44. 


group than the Poles as a whole. In other words, there was a definite 
selection of a special physical type which influenced some Poles to come to 
America and others to stay at home. Dr. Shapiro has found that the Japa- 
nese who migrated to the Hawaiian Islands are significantly different in 
many metrical and morphological characters from their own relatives 
who remained at home. 10 This was determined not by a study of repre- 
sentative samples, but by the actual measurement of relatives, in Hawaii 
and Japan. 

In the same sense, the Americans of colonial British ancestry are not 
like Englishmen in the larger sense of the word. The English who went 
to America in the Colonial period were a definitely selected group 
selected on the basis of religion, social and economic position, and geo- 
graphical distribution. Once in America, under new conditions, com- 
parative isolation, and the intensive cross-breeding of relatively few family 
lines, this differentiation was accentuated. Once the arable lands of New 
England and New York State had been cleared and cultivated, the farmers 
who moved westward into the fertile Ohio Valley, and on successively to 
Indiana, Illinois, and Iowa, were not typical examples of the total popula- 
tion of which they were drawn. The selection of the mountain men of the 
Rockies, and of the early cattle rangers of the Plains was even more no- 

So far we have been considering selection in migration in reference to 
the new country settled by the immigrants, but this selection, when the 
migration occurs in any numbers, has an equally important racial effect 
upon the old country. The depopulation of Ireland and Sweden through 
emigration to America must have affected the racial constitutions of these 
countries, just as the mass exodus of several hundreds of thousands of 
Germanic tribesmen in the Volkerwanderung period must also have af- 
fected northern Germany and Scandinavia. 

A lesser selection in scope, but equally important in principle and in 
effect, is the selection of urban populations from rural sources. Numerous 
European studies have made it clear that the young men and women 
who leave their villages to seek a new manner of living in the cities are 
racially atypical of the village populations as a whole, and that the drain- 
age of these people from the fecund rural districts into the relatively in- 
fertile cities has a selective value in the determination of the physical 
nature of the rural population. 11 

Selective differences in emigration and immigration exist in the cul- 
tural as well as in the racial sense. The Corded invaders who moved 
westward into Europe did not carry all the trappings of Asiatic and south 

10 Shapiro, H. L., SM, vol. 45, 1937, pp. 109-118; also, Migration and Environment. 
u Rryn and Schreiner, Somatologie der Nonveger, pp. 342-344, will serve as an example. 


Russian culture with them; they took only those objects which they would 
find useful in their new environment, and easy to replace from local ma- 
terials. In the same way the early American plainsman and trapper did 
not fill his knapsack with lace sleeves, wine glasses, and silver shoe buckles, 
but carried only such clothing, weapons, and other equipment which he 
knew would be of service to him. Later on, after he had settled the new country, 
the more luxurious trappings of the old culture could follow, provided that he had 
maintained contact with his original home. 

This last principle again applies to race as much as to culture. The 
settlers who come to a new country later, after the ground has been ex- 
plored, are often drawn from a different segment of the original society, 
and may represent a different racial entity, with different cultural associ- 
ations and aptitudes from that of the pioneers. 

Having dealt with amalgamation and selection, there remains the prin- 
ciple of environmental response. That human evolution has been going on ever 
since the initial acquisition of the distinctive human traits, such as speech, 
the use of fire, and the making of tools, cannot be denied. Man did not 
stop evolving once he became a man. We have seen that Pleistocene man, 
of whatever type, was more primitive in some respects than modern civi- 
lized man. The reduction of face size, and especially of the masticatory 
apparatus, is, for example, one of the most marked and most widespread 
active human evolutionary trends. 12 

There are other responses, however, which are not necessarily evolu- 
tionary, but which must be considered direct reactions to environmental 
change, in a broad sense. Changes in type and complexity of civilization, 
acting presumably through nutritional agencies, may serve as environ- 
mental stimuli and produce somatic effects. These responses, as observed 
in modern times, take the form of sweeping trends. The increase in stature 
which has affected northern and western Europe and much of the New 
World so profoundly within the last century is just such a trend. That it is 
a true mass response and not merely a selective process is shown by Bowles's 
studies of three generations of Harvard freshmen, taking only actual ge- 
netic lines of grandfathers, fathers, and sons. 18 That it is culturally moti- 
vated, whatever the mechanism, cannot be denied, for it is found only in 
countries which have been modernized progressively and thoroughly dur- 
ing this period. 

The most striking modern stature increase must be that of the English 
colonists in Queensland, for which there is ample evidence but no avail- 
able scientific data. The Queenslanders have shot up to an immense 
height, uniformly and with few exceptions, and have acquired a lanky, 

u Aahlcy-Montagu, M. F., QRB, vol. 10, 1935, pp. 32-59. 
" Bowles, Gordon T., New Types of Old Americans at Harvard. 


leptosome bodily habitus. Since the Queenslanders are essentially pio- 
neers, living largely off the soil, this must be due to direct environmental 
stimulation in the geographical sense. 

Stature increases may be matched with equally marked decreases. Dur- 
ing the Dark Ages, from the time of colonization to the sixteenth century, 
the Icelanders, originally as tall as their Norwegian ancestors, shrank in 
stature to the size of southern Italians. 14 Climatologists now tell us that 
this shrinking accompanied a lowering of mean annual temperature, and 
an increased dampness. 16 Icelandic history adds that it was a period of 
near starvation. The Greenlanders, who suffered even more from this 
climatic change, became even smaller than the Icelanders before their 
extinction. 16 Yet the Icelanders who survived this depression grew rapidly 
once it was over, until at present they comprise one of the tallest groups in 
Europe. The population of Iceland has not been materially added to by 
migration since the initial settlement. 

One of the best examples of environmentally conditioned physical 
stunting is to be seen in the misery area of the Limousin hills in central 
France. 17 Here isolation, poverty, and the dependence on the produce of 
an infertile granitic soil seem without reasonable doubt to have been the 
contributing causes. Mineral deficiency, in the sense in which Marett uses 
it, may be invoked, as well as malnutrition. Another example of environ- 
mental conditioning may be seen in the common level of short stature, for 
the most part below 160 cm,, which extends in a circumpolar zone around 
the world. 

If environment can so demonstrably affect stature, and act with such 
rapidity (the New Englanders have grown 7 cm. in 100 years), then it is 
more than likely that it can affect other racial criteria, including head 
form. The excessive brachycephalization which swept over central Eu- 
rope in the Middle Ages, affecting especially southern Germany and 
Bohemia, followed the same pattern as the stature change. Both pro- 
ceeded as orderly increases at fixed rates of speed. Selection may have 
been a large contributing cause, through infiltration and differential birth 
rates. Simple Mendelian dominance of brachycephaly, which has never 
been demonstrated, may not be eliminated, but it cannot have been the 
only factor involved. But even if we grant infiltration and differential 
selection. and direct Mendelism, it is difficult to account for the rise in 
cephalic index in the south German and Alpine region over any level 

14 Seltzer, G. C,, unpublished MS. in Peabody Museum. Author's permission. 
"Brooks, C. E. P., QRMS, vol. 47, 1921, pp. 173-190. 
* 8 Hansen, Fr. C. C., MOG, vol. 67, 1924, pp. 291-547. 

"Ripley, W. Z., Races of Europe, pp. 168-171, after Collignon, R., MSAP, ser. 3, 
vol. 1, 1894, pp. 3-79. 

Collignon, R., AG, vol. 5, 1896, pp. 156-166. 


which it had attained in antiquity, historic or prehistoric, unless we place 
this change at least partly on the basis of a response to environmental 
stimuli. The food-gatherers of west-central Europe seem to have re- 
sponded to an earlier and equally extensive brachycephalization during 
the Mesolithic, a period of profound climatic change; and the parallel 
modification, millennia later, among civilized food-producers, may, for 
reasons as yet unknown, have followed a parallel mechanism of change. 

All of this leads us back eventually to where we started, when we began 
to consider the meaning of the word race. A race is, in view of this discus- 
sion, a group of people who possess the majority of their physical charac- 
teristics in common. A pure race, if the term need be used, is one in which 
the several contributing elements have become so completely blended that 
correlations fail to reveal their original combinations. 18 At the same time 
the processes of selection and of response to environmental influences have 
given the resultant blend a distinctive character. 

The longer such a human entity remains isolated, the more distinctive 
it may become in the racial sense. It may expand numerically, divide, and 
become a major human stock, while others once much more numerous 
may become almost extinct, or fully so through absorption. But the most 
important fact about a race is that it is an entity, however ill defined, which 
is never static, but always in process of change. 

If, as above, we define race as a group of people reasonably unified in the 
physical sense and living in one place, difficulties at once arise. How are 
we to draw the borderline between that place and the next? Where does 
one race leave off and the next begin? There are those who assert that a 
race is merely an artificially assumed point on the smooth and glassy sur- 
face of a geographical continuum, 19 for what may be the concentration 
point for an extreme condition in one criterion will be an intermediate 
point in others. This assertion is, to a certain extent, true. If we view the 
panorama of living races on a two dimensional map, we can but agree that 
a race in this sense is merely a reasonably homogeneous group of people 
who occupy a given arbitrary point upon a terrestrial continuum. In 
regions of geographical smoothness one condition blends broadly and 
gently into another; in regions cut up by geographical barriers, such as 
deserts or mountains, the contrasts are sharper and the transitions more 

As long as we confine our glance to the surface, we will continue to be 
faced with this dilemma. But a solution comes with the application of a 
third dimension, that of history. By means of an historical reconstruction, 
with numerically adequate and competently documented skeletal material, 

Scheldt, W., ZFMA, vol. 27, pp. 94-116. 

19 1 am indebted for this concept to Dr. George Woodbury. 


it should be possible to determine what has happened in most regions 
occupied by the white race; why present conditions obtain; and what is 
a suitable classification of existing races, built upon the dual basis of the 
past and present. 

This classification must, of course, meet existing conditions and not 
be an expression of history alone, or of national ideals. By means of such 
a classification we may hope to answer the continuum objection, and show 
which spots on the map do actually represent centers of racial dissemination 
and which have functioned more characteristically as zones of inter- 
mediacy and blending, in accordance with principles which are now be- 
ginning to be understood. 20 

But we must remember, at the same time, that zones of intermediacy 
and blending may change their function without warning and assume the 
r61e of feeders of racial material to other regions. The interplay of these 
functions, in accordance with the principles already detailed in this chap- 
ter, has produced the racial complexity which characterizes most of the 
earth, and especially those portions occupied by the more active and vigor- 
ous and numerous branches of man, the Negroids, the Mongoloids, and 
the Whites. 


The materials used in the racial study of European man divide them- 
selves naturally into two classes: (A) skeletal material, including crania, 
long bones, and other bones such as vertebrae, pelves, tarsals, etc.; and 
(B) measurements and observations taken on the living. Both are subject 
to statistical treatment; and both must be employed if we are to succeed 
in our attempt to trace the racial history of white humanity. In the next 
six chapters, we will deal almost exclusively with material of the first cate- 

Museums both public and private, in almost every European country 
as well as in America, contain thousands of crania and long bones which 
represent the osseous remains of individuals of every race. Many of these, 
without doubt the majority, are those of persons of white racial origin. 
For the purposes of the present study, these skeletal remains assume vastly 
different values, depending upon a number of circumstances. In the first 
place, only those which have been measured, described, and published 
were of any use to the present author, since it has not been possible for 

Ketter, F., ZFRK, vol. 3, 1936, pp. 40-46. 

21 For an exhaustive study of this subject the reader is referred to the standard text 
in physical anthropology, Rudolf Martin's Lekrbuch der Anthropologie, 3 vols., second 
edition. The present section is intended merely as a brief statement concerning 
some of the fundamental uses of osteometric techniques, as well as of the sources and 
numbers of materials, employed in the present study. 


him to travel from museum to museum measuring and observing the 
unpublished material. The majority of collections are still unpublished, 
and hence the majority of data is as useless as if they were still in the 
ground. To make such a measuring trip would probably take the best 
years of one investigator's lifetime. 

The first consideration is, then, whether or not the material has been 
published. The second is, whether or not it is properly documented as to 
sex, provenience, and cultural association. A number of older cranial 
series has been published without regard to sex, which makes measures 
of variability of slight value, and jeopardizes the use of means. Others in- 
clude skulls from different localities, vaguely labelled and catalogued, 
which should never have been put together. Still others, and these are 
many, were unearthed at a time when the archaeologists had not yet so 
perfected their techniques that the cultural and chronological associations 
of these remains could be determined. Still others were brought into 
museums by amateurs who paid no attention to archaeology. 

In many cases it is possible to review the published documents as to 
archaeological settings, and to revise them in the light of present knowl- 
edge, especially when illustrations are given identifying the grave furniture 
and types of sepulchre. Therefore the number of crania and other bones 
which may be realigned so as to fit into geographical, cultural, and chrono- 
logical pigeon-holes is not as small as it might be if this material were 
gathered without recourse to this salvaging process. The realignment 
mentioned above is the principle upon which the following six chapters 
have been constructed. It has involved abstracting single skulls and small 
series of crania, with or without accompanying long bones, and combining 
the data so abstracted into statistical series based on an identity of place, 
time, and cultural milieu. In some cases earlier investigators had already 
effected this process of compiling and combining in a suitable way, so 
that much of the labor involved could be omitted. 

The materials upon which Chapters II to VII are based consist, there- 
fore, of a number of series of crania, in some cases accompanied by other 
bones, each series representing a cultural, chronological, and geographical 
entity, the existence of which seems fully justified in the light of our pres- 
ent knowledge of archaeology and of history. Published materials which 
cannot be reasonably documented in all of the respects mentioned have 
been ignored, or used with caution. 

The crania which meet these requirements and which represent an- 
cestral strains of the white race are numerous enough to permit a reason- 
able reconstruction of the racial history of the white peoples; but they are 
not numerous enough to permit us to be sure that our reconstruction is 
the only possible one in every place and instance. We therefore present 


with some confidence the main thesis of our reconstruction, but we are 
not confident that it is correct in every period, in every region, and in every 
cultural unit. 

The entire Palaeolithic period in Europe, for example, is represented by 
no more than one hundred published and documented skulls, while the 
Mesolithic is represented by a no greater number. Certain Neolithic 
samples, especially in Egypt, consist of several hundreds of crania, and the 
same is true in the Bronze and Iron Ages. No craniological series yet 
published exceeds one thousand adult specimens of a single sex, although 
several closely approach that figure. 

Skeletal material of human and near-human primates, from the Lower 
and Middle Palaeolithic cultural levels, is derived from chance finds of 
unburied fossil bones. In Europe, Neanderthal man first buried his dead 
in such a way that entire skeletons would be preserved for anthropologists 
of the future. At various points in human history cremation appeared, to 
confuse and dismay the racial historian; the chief vogue of this science- 
inhibiting custom began during the late Bronze Age in Europe, and lasted 
well into the Iron Age. 

In our era another force has arisen to prevent the use of skeletal ma- 
terial; this is the practice of burying bodies in Christian and Moslem 
cemeteries, both of which are inviolate on religious grounds. Even where 
they are not inviolate, the absence of grave furniture in the tombs of 
these followers of revealed religion makes looting by archaeologists un- 
profitable. The only skeletal collections of any abundance in post- 
Christian times are those derived from mediaeval charnel houses or crypts, 
especially in South Germany and Austria, and in certain English cathe- 

From the statistical standpoint our skeletal materials stand in a border- 
line position. A few series are large enough to permit the exercise of all 
of the statistical constants of the modern biometric school; most, however, 
are so restricted in numbers that a simple calculation of mqans, a simple 
determination of variability and homogeneity, and an informal compari- 
son and discussion are the only techniques which seem justified. 22 Too 
great a mechanization would render such series inflexible and destroy 
much of their interpretive value. To make up for their statistical weakness, 
their use as context material for cultural and chronological horizons pro- 
vides a certain strengthening. A series, however small, tells us what is 
present, but does not tell us what is additionally present, or what is absent. 
The extent to which small series may be employed in an interpretative 
sense must depend upon the circumstances. 

22 For a more detailed discussion of the use of statistics in racial studies, see Chap- 
ter VIII, section 2. 


The number of criteria measured upon the crania used in this survey 
range from one almost always the cranial index to the five thousand of 
von Torok. In combining and reseriating these series I have made no fast 
rule as to what criteria to admit and what to exclude, but have employed 
what seemed to be a reasonable number, with especial emphasis upon 
those which find parallels on the living. For example, I have usually ac- 
cepted the three principal dimensions of the cranial vault glabello- 
occipital length, maximum biparietal breadth, and basion-bregma height; 
the usual circumferences and arcs of the cranial vault; the minimum and 
maximum frontal and bizygomatic diameters; the interorbital and bi- 
orbital diameters, and the height and width of the orbits; the height and 
breadth of the osseous nose, the diameters of the palate, and of the fora- 
men magnum; the heights of the face from nasion to men ton, and nasion 
to alveon; the principal dimensions of tfce mandible, such as the mental 
height, the breadth of the ascending ramus, and the bicondylar and bi- 
gonial diameters. In the rest of the skeleton, I have used almost exclu- 
sively the maximum lengths of the principal long bones, such as the femur, 
tibia, fibula, humerus, radius, and ulna, and then almost entirely for the 
sole purpose of reckoning stature, by means of the Pearson formulae. 23 

In other words, I have used what I could find in such a way as to de- 
rive the maximum useful information from it; I have not concerned my- 
self with techniques or routines which had little bearing on my problem. 
On the whole I have worried little about technical discrepancies due to 
differences in measuring methodology; where possible I have followed the 
techniques approved by Morant, and where possible I have made allow- 
ances for such differences as I could readily detect. I do not feel, however, 
that technical discrepancies in the craniological materials are important 
enough to make any perceptible difference in my conclusions, either de- 
tailed or general. The treatment of the material has been done in such a 
broad manner that such minutiae are of little importance. Craniology is 
a more accurate science than is the anthropometry of the living; when 
we come to the later chapters we may concern ourselves with the question 
of technique, but for the moment it is relatively unimportant. 

23 See Martin, Lehrbuch der Anthropologie, second edition, vol. 2, pp. 1020-1021. 

Chapter II 


Man as we know him from the study of modern races is descended 
from one or more species of a single genus, segregated out of a group of 
related Old World primate genera which had physically, and hence cul- 
turally, taken the first definite steps in a human direction. Members of 
a number of these genera found that they could cut with the sharp, glassy 
edge which is formed when flint is fractured, learned the use of fire, and 
discussed these and other matters with their fellows by means of speech. 
But all of them, including those destined to take part in the formation of 
the modern species Homo sapiens, remained, like less human primates and 
other animals, dependent on the natural occurrences of foodstuffs for 
their continued existence. 

Man alone, of these parallel forms, succeeded in breaking loose from 
the natural limitations of food and climate, and he did this in a number 
of different ways. One of these was the invention of warm clothing which 
would permit him to hunt in comfort the numerous arctic and temperate 
mammals, whose flesh was richer in fats than the meat of his tropical 
prey; but this was not immediately a greater advantage than the develop- 
ment of a furry coat among the animals which he hunted. Another was the 
discovery of the principles of reproduction in animals and plants, and the 
knowledge of how to control this reproduction. This second step, which 
Childe calls the first revolution, produced of course agriculture and ani- 
mal husbandry, and out of this dual economy have developed the civi- 
lizations of the ancient and modern worlds. 

These two primary steps, which were in no sense consecutive, although 
the first was undoubtedly the earlier, and which had no necessary relation- 
ship one to the other, brought about different effects of far-reaching con- 
sequence. The first of them permitted the utilization by man of lands 
which could not otherwise have supported primate life, and also the 
ability to pass through the arctic barrier from the Old World into Amer- 
ica; the second, the intense use of more favored regions, suited for farming 
and pasture, and the increase in population made possible by the conse- 
quent abundance of foodstuffs. These two steps, then, permitted the 
human species to multiply greatly, and to occupy all of the principal inter- 
connecting areas of the earth, not covered by glaciers or waterless deserts, 



and not separated from the mainlands of the two hemispheres by wide 
expanses of sea. 

Long before either of these two steps had been taken, most of the re- 
lated primate genera and species, which had participated in the earlier 
discovery and utilization of speech, flint, and fire, had dropped out of 
the contest. Perhaps the last to disappear was Homo neanderthalensis, who 
became extinct as such in Europe, if not elsewhere, at the time of the last 

Homo sapiens, then, as we now know him, remained alone to deal with 
the results of the increasing control over nature which he himself had con- 
jured. But all branches of the species did not participate in these results, 
while tho'se which have participated have not shared them equally. In 
the far peripheries of the southern hemisphere, to which man in the 
earliest stages of culture could retire without encountering great cold, 
naked hunters and gatherers, such as the African Bushmen, the Tasm'a- 
nians, the Australians, and the Vedda, have been able to survive in isola- 
tion until recent years. While only one, the Tasmanian, is extinct in the 
unmixed form, 1 the others promise soon to follow. Thus our species is 
repeating within its own ranks the selective process of elimination which 
effaced, in earlier times, its non-human competitors. 

There is, actually, no real difference between these two cycles of ex- 
tinction. In the first, although separate species were involved, we now 
know that at least one of those which disappeared was not pruned off 
the stem completely; for, as in the second cycle, its disappearance was 
consummated by absorption coincident with cultural changes, permit- 
ting the submerged genetic strain to survive in solution. Human ge- 
netic strains, however ancient and however primitive, are very hard if 
not impossible to eradicate completely, for the simple reason that 
all human racial stocks are mutually fertile, and men of all races are 

Nevertheless these racial stocks possess, under varying conditions, very 
different rates of procreative value. It is a constant phenomenon of 
human history that a small group of people in a restricted area will, 
through some stimulus which is probably both environmental and cul- 
tural, increase rapidly, expand its boundaries, and inundate new seg- 
ments of the earth's surface with its progeny. For example, the numerical 
size of the white race has, since the time of the industrial revolution, in- 
creased vastly. The countries in which the new regime was initiated grew 
much more rapidly in population than did those yet to acquire these cul- 
tural innovations. In this manner, emigrants from Europe spread out 

1 A few mixed survivors live on the islands between Australia and Tasmania, and in 
reservations on the Australian mainland. 


into other continents, previously occupied by less economical 2 popula- 
tions, until they and their descendants had filled most of the available 
space suited to their powers of utilization. After this, their rate of increase 
fell. New conditions and new stimuli, provided that they are favorable, 
produce great increases in a species. Unfavorable ones produce absorption 
and extinction. 

This phenomenon is not confined to human beings, but is a basic prin- 
ciple of biology, by which has been accomplished the spread of all plant 
and animal forms. Man, whose ancestors were a handful of precocious 
and biologically successful primates, has multiplied until his numbers are 
now reckoned in billions. The present numerical proportions of races 
and of nationalities has no reference to the former numbers of previously 
existing groups of people, to the importance of these various groups in the 
history of human racial development, nor furthermore, to their relative 
numerical values in the future. In the subsequent pages outlining the 
racial history of the white segment of mankind, this principle must not 
be forgotten. 

This history is, on the basis of present knowledge, entirely confined to 
Pleistocene and Recent geological time. It is with the earlier of these two 
segments of the Genozoic 3 that the present chapter is concerned. In it, 
as in subsequent chapters, some attempt will be made to place the skeletal 
remains studied in their proper chronological horizons. With the Pleisto- 
cene specimens, this dating must be done primarily by geological means. 

Since the primary diffusion of a zoological species is almost instanta- 
neous, palaeontologists base their dating of geological horizons on the 
initial appearance of fossil genera and species. By this means they have 
divided the Pleistocene period into lower, middle, and upper levels. 4 
Glacial geologists, limited to the relatively small portion of the earth's 
surface which was covered by one or more of the Pleistocene ice sheets, 
divide it by reference to the four or five successive glacial advances. Un- 
glaciated regions were subjected, during the Pleistocene, to alternations 
of wet and dry climate, probably correlated with the vacillations of the 
ice. The pluvial and interpluvial periods so determined form a third 
means of dating Pleistocene remains. At the moment, a complete har- 
mony between these three systems has not been achieved. Hence the 
relative dating of fossil human beings found in various parts of the world 
is not, as yet, wholly possible. For that reason we must proceed with 
reserve and caution. 

2 In the sense of maximum utilization of the soil. No qualitative inference is intended. 

3 The division between Pleistocene and Recent is here maintained purely for the 
sake of clarity. The possibility that we are now living in a Pleistocene interglacial is 
not, by the use of this terminology, implicitly denied. 

* Hopwood, A. T., PGA, vol. 46, 1935, pp. 1, 46-60. 



It is not easy to overemphasize the importance of climate in human 
history, particularly in the earliest times when man was merely a numer- 
ically unimportant parasite in the total fauna. With changes in climate, 
he was forced to migrate with the animals and plants on which he lived, 
and at the hunting and gathering of which he was adept. The only 
alternative was to stay on and adapt his culture to a new food supply, 
which would need new implements and new methods. On the whole, 
it was easier to move, even if some of the oscillations were, like those in 
recent times, rather rapid. 

The ponderous ebb and flow of the glaciers caused climatic changes 
which affected the entire world. With the gathering of vast quantities of 
ice near the poles, zones of climate shrank inward, converging on the 
equator. At times of maximum glaciation, wide belts of land bordering 
the glaciers became treeless, frozen tundras, like the northern rims of 
Siberia and North America today. During the last glaciation, such a zone 
included the whole of Europe north of the Alps and Pyrenees, and much 
of Siberia. Below this stretched temperate forests, with zones of willow 
and birch, of pine, and of hardwood, and beyond these, temperate, grassy 
plains, watered by cyclonic rain belts. Still farther away, near the equator, 
stood tropical forests. The present deserts had shrunk to narrow patches 
between the grasslands or had disappeared. 

As the glaciers retreated, the zones of tundra followed, constantly 
shrinking as the ice cap thinned. The forest encroached on the tundra 
belt, and the grasslands likewise moved inward; at the same time the 
tropical forest shrank, and the land in between two belts of grassland 
became desert. What had once been the optimum home for food gather- 
ing man now became bare and sterile, and remained virtually unoccupied 
until the rise of pastoral nomadism, with ass and camel, once more made 
it habitable. 

The centers of Pleistocene glaciation were not located exactly on the 
poles. In the northern hemisphere, the center was in the north Atlantic, 
with land nuclei in Scandinavia, northern Britain, and Greenland, so 
that northwestern Europe and northeastern America were covered, while 
territories of higher latitudes, in eastern Europe and Siberia, and in 
western North America, were left bare. In Europe, the ice covered, at 
its maximum, all of the British Isles but the southwestern tip of Great 
Britain; most of Belgium, Holland, northern Germany, the Baltic States, 
and Finland, as well, of course, as Scandinavia. Secondary centers of 
glaciation, based on altitude rather than latitude, lay in the Alps, Pyre- 
nees, and Caucasus, in the Himalayas and Pamirs, in the mountain skele- 
ton of Siberia, and in the Atlas mountains of North Africa. 


These ice caps, and the surrounding zones of cold, acted as barriers to 
the naked hunters of the Early and Middle Pleistocene. In Europe, no 
sure instance has been established of a Lower or Middle Palaeolithic find 
in a glacial context; before the first Wurm glaciation, human beings and 
related primates gave the ice a wide berth. 

During the entire span of the Pleistocene up to the fourth or Wurm 
glaciation, bands of human beings, probably including both sapiens and 
non-sapiens forms, shifted slowly from continent to continent with the 
changes of climate. During the fourth glaciation, the parts of Europe 
and Asia immediately south of the ice sheet, and in the tundra belt, were 
for the first time, under such conditions, inhabited. This was by Neander- 
thal man, who lived in caves, warmed himself over fires, and could, 
judging by his tool kit, dress skins, although, in default of needles, he was 
probably a poor tailor. The European branch of this species was a mar- 
ginal, primitive form, and barely survived the 'fourth ice. During the 
Laufen interglacial, Neanderthal was replaced in Europe by pure and 
mixed sapiens men coming from the east in several waves. With the last 
major ice advance, Wurm II, sapiens man stayed on, for by now he had 
developed the knowledge and skill to make warm clothing, as numerous 
skin-working tools and fine bone needles attest. 

In the meanwhile, other sapiens men must have lived in more favorable 
climates, as much on vegetable food as on meat. Some of these developed 
the microlithic cultural technique, which involved striking off small 
blades for composite instruments, and this spread to Europe north of the 
Pyrenees only after the retreat of the last ice. These sapiens men were, 
as we shall see, quite different from those in the North. The post-glacial 
movements of human groups completely changed the racial complexion 
of much of the habitable earth. 


The first appearance of fully or incipiently sapiens men in the Old 
World can now be definitely placed in the Middle Pleistocene, in Europe 
the time of the second, or great, interglacial. The specimen which has 
made this allocation possible is Swanscombe man, consisting of a parietal 
and the occipital bone of one individual from a glacially sealed Middle 
Acheulean deposit on the second terrace of the Thames Valley in England. 6 
These fragments are said to resemble the cranial vault of Piltdown, which 
is also probably sapiens in the same sense, and may be of no greater antiq- 
uity. 6 

8 Swanscombe Committee of the RAI f JRAI, vol. 68, 1938, pp. 17-98. 
See especially Morant, G. M., ibid., pp. 67-96. 

8 It is becoming increasingly unlikely that the Piltdown mandible is a part of the same 
specimen as the vault fragments. 


Other remains comparable to those from Swanscombe, and also asso- 
ciated with the Acheulean cultural horizon, have been found in various 
sites in western and southern Europe, but have so far failed to receive full 
scientific recognition. The best known of these is the famous Galley Hill 
skeleton, found in the second or hundred-foot terrace of the Thames 
Valley. Others include the Moulin Quignon mandible, the Clichy skele- 
ton, and the Olmo skullcap. Of these, the most nearly complete, 7 and the 
strongest claimant for authenticity, is the Galley Hill skeleton, unearthed 
in 1888. 8 Although the skeleton was removed from near the bottom of an 
undisturbed gravel layer, by persons fully aware of the importance of its 
position, most modern writers of 'the pre- Swanscombe era have refused 
to accept its authenticity, although the chances of its being later than the 
gravel from which it was taken were at most extremely slight. In view of 
the Swanscombe evidence the Galley Hill specimen may now be granted 
the recognition which it has long merited. 

The Galley Hill man was of short stature, about 160 cm. His long bones, 
which include a humerus as well as a femur and tibia, although robust, 
were not heavy. The length of the tibia is 77 per cent of that of the femur, 
and this proportion is modern and European, unlike those of many of the 
later peoples of the Upper Palaeolithic and Neolithic. From the muscu- 
lar markings on their surfaces, it is apparent that he was a man of con- 
siderable bodily strength, but at the same time of fairly light build. The 
section profiles of the long bones, the positions of the condyles, and the 
facets, all bear witness to a life in open country, and to the habit of squat- 

The skull, which is reminiscent in a general way of some living varieties 
of European man, is extremely dolichocephalic, with a cranial index of 
69; although warped by earth pressure, it has not changed its basic form. 
The length of the vault is very long, 204 mm. as reconstructed by Keith, 
and the breadth correspondingly narrow. The vault height, known only 
from the auricular projection since the Ipasal portion of the skull is missing, 
is on the low side of medium. This skull has an extremely protuberant 
occiput with the greatest length well to the bottom; a well-developed 
frontal region, and a moderately sloping forehead. At the same time the 
forehead is very broad, making the parietal walls nearly parallel. The 
browridges are of moderately strong development. The face, unfor- 
tunately, is missing in Galley Hill as in all similar specimens. Yet the 
temporal segment of the right zygomatic arch remains, and this, although 
thin, shows that the arch as a whole was well curved. 

Fortunately, more than half of the mandible has been preserved, and 

7 The Clichy skeleton may be more complete, but has not been satisfactorily published. 

8 Keith, Sir A., The Antiquity of Man, pp. 178-193. 


its conformation makes it certain that there was no prognathism. The 
body of this mandible is rather narrow and of only moderate symphysial 
height; the chin of medium prominence judged by modern standards. 
The ascending ramus is wide, and the sigmoid notch shallow. The teeth, 
while fully human, retain some primitive features in the development of 
the pulp cavities, in the length-breadth proportions of the molars, and in 
their relative size, for the third molar is the largest. 

Besides these dental peculiarities and the absence of a marked sigmoid 
notch, the skull itself possesses certain primitive features. It is thick, and 
the browridges, although no greater than in many modern examples, 
form a continuous ridge. The mastoids are small, and the area of tem- 
poral muscle attachment large. 

Galley Hill man was, without reasonable doubt, an extremely gen- 
eralized form of ancestral white man. His skull and body bones preserve 
just that degree of generalization needed to make him the logical ancestor 
of the Mediterranean race and of all the sub-races related to it. 

Although more specimens of this type have so far been found in Europe 
than elsewhere, it is not possible to suppose that the Galley Hill type of 
man evolved on European soil. He must have been a transient in Europe, 
coming in with the retreat of one glacier, and going out again with the 
advance of the next. When his descendants next appear in Europe, it will 
be from some other source to which their ancestors had retreated. 

Outside of Europe, the earliest known human anatomical specimen 
is the Kanam mandible from East Africa. This was attributed by its 
finder, Leakey, 9 to the Lower Pleistocene, which would probably make it 
older than any of the other known fossil men of Africa, Asia, or of Europe. 
The Kanam mandible is definitely human; it possesses a chin and its 
teeth are essentially human in form, although primitive in a number of 
ways, 10 like those of Galley Hill. It is impossible to determine with any 
accuracy the racial type represented by this fragment of jaw; especially 
since, if it possesses the age attributed to it, races in the modern sense 
cannot have developed very far. However, it could, like Galley Hill, have 
belonged without difficulty to a generalized ancestral white man, since 
it lacks prognathism and is modern in shape and size. 

Younger than the Kanam mandible, and apparently belonging to the 
Middle Pleistocene, are four fragmentary skull caps found, likewise by 
Leakey, in East Africa at the site of Kanjera. These, like the Kanam jaw, 
have been subjected to the investigation of a British Committee which is 
not satisfied as to the exact location from which they came. However, as 
Hopwood has pointed out, the fossils from both the Kanam and Kanjera 

Leakey, L. S. B., The Fossil Races of Kenya. 
"Adloff, P., ZFRK, vol. 3, 1936, pp. 10-26. 


deposits belong to the periods which Leakey stated; namely, the Lower 
and Middle Pleistocene. 

Despite the uncertainty of this situation, in view of their great impor- 
tance, and the fact that their alleged age has not been disproved, it would be 
unwise to ignore these East African specimens in a theoretical reconstruc- 
tion of the history of Homo sapiens. It is much more reasonable to give them 
full consideration and to label the sequence of reconstruction as tentative. 

These four fragmentary skull caps found at Kanjera are in such poor 
condition that it is impossible to give accurate measurements or other 
details which would fully define the types which they represent. Yet 
enough pieces have been preserved to make a general estimate. Kanjera 
man was extremely dolichocephalic, with cranial indices under 70; the 
skull walls, although thick in three out of four cases, are not covered with 
heavy muscular markings, as in the case of non-sapiens types of fossil man. 

The foreheads are prominent; the frontal lobes of the brain well de- 
veloped, as in any modern group; the whole occipital region is extremely 
protruding, and the occipital lobes strongly developed and very sym- 
metrical. This fact, along with other features of the brain deduced from a 
study of the endocranial casts, 11 leads one to the conclusion that these 
specimens belonged to a very long-headed form of Homo sapiens, very 
similar to Galley Hill, and like the latter could without difficulty have been 
ancestral to at least one part of the present white racial stock. One small 
piece of malar bone is all that remains of the faces of these four individuals. 
This fragment includes a well-developed canine fossa, which again is cer- 
tain proof of its human character. A small piece of femur with a strongly 
developed pilaster is also fully human, but cannot serve to designate any 
single racial group. 


Not demonstrably older than the Pleistocene fossil men discussed in the 
last section are the remains of an increasingly large number of non-sapiens 
specimens from all three continents of the Old World. 12 These include 
two separate genera, Pithecanthropus and Sinanthropus, and four species of 
Homo soloensis, heidelbergensis, neanderthalensis, and rhodesiensis. The exact 
relationships between these groups is in dispute, but it is apparent that 
they may be grouped in at least two evolutionary levels, with Pithecan- 
thropus and Sinanthropus in the lower bracket. Despite their allocation to 
separate genera, these two are, in many respects, very much alike. Fur- 
thermore Rhodesiensis and Soloensis resemble each other, and together 

Elliot-Smith, Sir G., The Stone Age Races of Kenya, Appendix B. 
11 The extensive literature on these fossil groups need not be cited here. Except in the 
case of Neanderthal, they have little bearing on the subject of this book. 


are not very different from the numerous and variable Neanderthaloid 

These fossils, whatever their internal classification, may be considered a 
separate class of highly evolved, humanoid primates. Within this class 
there are differences of evolutionary status, and differences in type of 
specialization. As a whole, however, they differ from both early and 


MacGregor's restoration of La Chapelle aux Saints, provided with hat, hair, and 
clothing by the artist. Although we do not know that the reconstruction of the soft 
parts is accurate, nevertheless the facial features were probably essentially human. This 
picture serves to illustrate the fact that our impressions of racial differences between 
groups of mankind are often largely influenced by modes of hair dressing, the presence 
or absence of a beard and clothing. 

modern sapiens man in the possession of a flattened, gorilloid skull vault, 
with a strong supraorbital torus, an extremely sloping forehead associated 
with a low vault height, and a strongly girded brain case, in which, in the 
more primitive species, the maximum cranial length passes from glabella 
to an occipital torus, while the maximum breadth lies between the mastoid 
crests. Even in the more evolved species in which the brain size equals or 
exceeds that of modern men, the same gorilloid structure to a large extent 
persists. The faces of the few specimens which still possess them are of 
extreme length and breadth, and the subnasal portions excessively large 
in comparison to the brain case; these faces are flat, and that distinctive 


human feature, the canine fossa, is lacking. In the case of most known 
Neanderthals, the molar teeth have fused roots and enlarged pulp cavities, 
while the dental borders are even, and the canines not interlocking. 18 

The dating of the various fossils mentioned above is in most cases under 
dispute, but there is no valid evidence that any of them are earlier than 
the Middle Pleistocene. Only Homo neanderthalensis in some of his more 
highly evolved forms is known, however, to have extended into the Late 
Pleistocene. Aside from all biological considerations, the time element is 
sufficient to destroy the hypothesis that members of this heavy brow- 
ridged group could have evolved directly into the earliest known form of 
Homo sapiens. It is possible that these species represent a survival of an 
ancestral stage through which Homo sapiens had in earlier times passed, 
and that they were, during the Pleistocene, themselves passing through 
a tardy process of evolving, but this explanation is not the only one that 
may be presented. The sexual differentiation and luxuriance of gorilloid 
characters which these species possess may conceivably never have been 
found in the direct ancestor of sapiens man. 


In western Europe, Neanderthaloid skeletal material begins to appear 
in the second interglacial, with the Heidelberg jaw, 14 and is followed, dur- 
ing the early part of the Riss retreat, by the Steinheim and Ehringsdorf 
crania. The whole of the third interglacial, and the advance of Wurrn I, 
belonged to Neanderthal men, and not a single sapiens skull has been 
found, in Europe, dating from this long time expanse. 

The Neanderthal group was extremely variable, and showed within its 
ranks clear evidence of evolutionary change in a human direction. On 
the whole, the western European specimens formed a marginal, and rela- 
tively primitive, geographical sub-group of the species. The center of its 
dispersion probably lay farther east, as did, one may suppose, that of 
the Mousterian flake culture with which the Neanderthal species seems 
to be basically associated. 

In Palestine, which falls on a periphery of this cultural range, excavations 
in caves near the Sea of Galilee and Mount Carmel have revealed a num- 
ber of Neanderthaloid skeletons which are different from those in Europe, 

18 The condition known as taurodontism is not as uncommon as has been supposed, 
and can no longer be cited as an impediment to the relationship of Neanderthal with 
other types of man. For a discussion and bibliography on the subject of taurodontism, 
see Galloway, A., The Skeletal Remains of Mapungubwe, pp. 127-174, in Fouch6, L., 

14 Current scientific opinion in Germany tends to place Heidelberg in the first inter- 


and others which are, in fact, only partly Neandefthaloid. 16 The materials 
from the Mountain Carmel caves, situated in a late Middle Pleistocene 
setting, corresponding to the latter part of the third interglacial in Europe, 
were found imbedded in a breccia thick with Levalloiso-Mousterian imple- 
ments. It is with these late Mousterians, who showed atypical racial 
features, that we are at present concerned. 

In one of the Mount Carmel caves, that of Tabun, was found the skele- 
ton of a small woman, fully Neanderthaloid, and associated with it was a 
male mandible equal in size to that of Heidelberg, but possessed of that 
human feature, a chin. In a nearby grotto, the Mugharet es-Skhul, were 
the remains of a number of individuals, including three male crania suffi- 
ciently complete for reconstruction and measurement. A preliminary 
publication 16 of three of these skulls, and of the long bones of the same 
and other individuals, gives us a reasonably accurate idea of their posi- 
tion in the human family tree. Originally considered members of the 
Neanderthaloid species, they are now known to be fully human, although 
preserving a number of unmistakable Neanderthaloid characteristics. 

The leg bones of the Skhul people are long and slender, the femora 
heavily pilastered, in contrast to the Neanderthaloid form. The feet are 
fully human, but lack the reduction found in the middle phalanges of 
modern races, while the heels are short. The humeri arc likewise long 
and slender, the radii and ulnae straight, instead of being bowed as with 
Neanderthal man, including the Tabun female. The hands of Skhul men 
were broad and large. 

In the Skhul pelves, definite Neanderthaloid features appear; the entire 
structure is lower and narrower than those of most modern men. The 
Tabun woman's pelvis, on the other hand, is quite different from other 
Neanderthaloids, in the possession of a long, plate-like pubis, which is an 
ape-like character. The vertebral column of the Skhul men, while human, 
and possessing a lumbar curve of sapiens character, is short in the cervical 
region. The total height of the cervical vertebrae is only 55.7 mm., as 
contrasted with a mean of 68.4 mm. for modern man. Thus the Skhul 
men were short-necked, and in this respect possessed a Neanderthaloid 
trait. In comparison with Neanderthal man, the Skhul thorax was flat, 
while that of the Tabun woman retained the barrel-like earlier form. 
The ribs of the Skhul men are variable in cross-section; some are flat and 
ribbon-like, as in modern man, others are thick and rounded, as with 

15 Keith, Sir A., "A Report on the Galilee Skull,'* in Turville-Petre, F., Researches in 
Prehistoric Galilee. 

Keith, Sir A., and McGown, T. W., BASF, #13, 1937, pp. 5-15; also "Mount Carmel 
Man," etc,, Early Man t Phila., 1937, pp. 41-52. (Other notices superseded by the last 
two mentioned.) 

Keith, Sir A., and McCown, T. W,, BASF, #13, 1937, pp. 5-15. 


Neanderthal. The latter form is also associated with the Upper Palaeo- 
lithic European men, 17 whose relationship to the Skhul people will be 
treated later. The stature of the Skhul males was tall, ranging from 173 
to 179 cm., while that of the females, estimated from long bones, was 
short, 158 cm. The sex differentiation thus revealed is great. 

In the skull, Skhul man is definitely intermediate between the Neander- 
thal sCnd sapiens groups, but much closer to the latter, so that its inclusion 
in the living species cannot be denied. The skulls of the three males are 
extremely large. In length, they equal Galley Hill, but far exceed it in 
breadth; the vault height of two specimens, #5 and #9, measured from 
the ear holes, is equal to that of Galley Hill, but the third, #4, is as low 
as with true Neanderthals, while the extreme breadth of this specimen 
acts as a compensation, permitting a greater capacity than with the other 
two. In vault form, then, two are mainly sapiens, while one appears, 
from the measurements, to be largely Neanderthaloid. The capacities of 
these three skulls are 1588, 1600, and 1616 cc., respectively, much greater 
than those of Galley Hill or others of his type, and greater than those of 
most living men. At the same time, they exceed most Neanderthal figures. 
In brain size as in stature, Skhul man exceeded either Neanderthal or 
Homo sapiens as previously known. 

The best preserved and most complete specimen, #5, is a heavy, thick 
skull, with heavy browridges, which do not, however, attain a maximum 
Neanderthaloid development. The greatest length falls higher, in the 
rear, than with the Neanderthals; although the occiput is protruding, it 
is not conical in form, as with many Neanderthal specimens. The vault 
is well-arched, the lambdoid region slightly flattened, and the forehead 
no more sloping than in many modern sapiens crania. 

The face, while large, fails to attain the gorilla-like proportions of 
Neanderthal, and falls within the modern range in height and breadth. 
The interorbital distance is, comparatively speaking, great; the upper 
bonders of the orbits straight. Both tjie maxillae and mandible are of 
great size and robusticity, exceeding most modern specimens, and the 
alveolar prognathism is excessive. The mandible has, however, a fully 
human chin, and the teeth are, like those of the Tabun specimens, not 
taurodont. The palate, viewed from below, while large, is long in pro- 
portion to its breadth, unlike Neanderthal in which the breadth exceeds 
the length. The foramen magnum, like that of Neanderthal, is long and 

Although the anthropornetric position of the Skhul crania will be dis- 
cussed later in more detail, it is worth noting at the moment that in most 
characters capable of measurement the #5 specimen falls between Homo 

w Aichel, O., Derdeutsche Mensch, p. 30. 


sapiens, as exemplified by Galley Hill and later examples of the same type, 
and Neanderthal, as known from the totality of that species. 18 

Keith and McCown have demonstrated, beyond serious doubt, that the 
Skhul skeletons are intermediate between Homo neanderthalensis and Homo 
sapiens, and that Neanderthal must therefore be included among the an- 
cestors of modern races. Thus the opinions of Hrdlifcka, Aichel, 1 and 
others, expressed earlier on the basis of equally valid but less striking evi- 
dence, are at last, in one sense or another, substantiated. We now know 
that the Neanderthal strain did not become extinct, but passed over into 
the genetic stock of modern man. If this occurred once, it could have 
occurred a number of times. The field is now open to discover survivals 
of non-sapiens accretions in modern races in other parts of the earth. This 
privilege must, however, be used with caution. 


The next step is to examine the evidence which reveals the racial com- 
position of Upper Palaeolithic man in Europe. Until the discovery of the 
Swanscombe fragments, these were the earliest sapiens remains which were 
definitely datable to the satisfaction of all interested scientists, and im- 
mune to the doubts which had thrown all supposedly earlier finds into 
the shade. 

The first Upper Palaeolithic people, the bearers of the earliest phase of 
the Aurignacian culture, arrived in Europe during the middle of the 
Laufen interglacial, between the retreat of Wurm I and the advance of 
Wurm II. On the basis of accurate Scandinavian chronology, it is pos- 
sible to set the end of the Upper Palaeolithic in western Europe with more 
accuracy; 11,800 B.C. seems to mark a turning point, with the migration 
of the reindeer northward, and the first introduction of Mesolithic culture. 
In view of the present differences of opinions between geologists, it seems 
unwise to set even a tentative date for its inception. In any case, the time 
that elapsed during the Upper Palaeolithic must have provided ample 
room for change in some of the more fluid physical characters of a people, 
especially if they have been subjected to rigorous climatic conditions and 
specialized diets. 

We must not place too much importance on fine differences in stature 
as a means of determining genetic affinity or distance, especially over 
periods of tens of thousands of years. Head form, too, although it changes 
with much less speed than stature, for it is not directly concerned with 

18 Figures for the latter obtained from Morant, G. M., AE, vol. 2, 1927, pp. 376-377. 
" Hrdliftka, A., The Skeletal Remains of Early Man, MCSI, vol. 83, 1930. 
Aichel, O., Der deutsche Mtnsch. 


gross size, nevertheless responds to the stimuli which control it, and we 
must not be surprised if long heads have in some instances become round 
heads during the course of hundreds of generations. 

In studying the remains of Pleistocene and of post- Pleistocene man, 
therefore, we must be careful not to confuse characters which are of racial 
importance with progressive modifications which may occur, in response 
to cultural changes, within any group. Such modifications are especially 
concerned with the jaws and teeth. Among the more primitive white 
peoples, such as Berbers and Albanian mountaineers, the incisors of the 
two jaws meet edge to edge, as they did among most of the mediaeval in- 
habitants of western Europe. Under modern conditions this changes 
rapidly to an overbite, and is frequently accompanied by a narrowing of 
the palate and crowding of the teeth, making modern orthodontia profit- 
able. This shifting of the bite affects also the position of the lips and 
changes the entire facial expression. 

Another modification which seems to proceed with some rapidity is the 
enlargement of the masticatory muscles under sub-arctic conditions. As 
these muscles enlarge, the angles of the lower jaw become everted, the 
zygomatic arches expand laterally, and often the brain case becomes 
keeled in response to an increase in temporal muscular attachment. At 
the same time both the mandible and the palate develop tori. These 
correlated changes act without regard to race since they are apparently 
functional adaptations. They also act with some rapidity, for the mediae- 
val Icelanders acquired them in less than four centuries. 30 These occur 
in varying degree among some of the later Upper Palaeolithic European 
skulls, as well as among Eskimos and modern Siberians. 

We must be particularly careful, therefore, in studying the remains of 
Upper Palaeolithic man, to remember that his time span was unques- 
tionably greater than the totality of time which has elapsed since it ended. 
We must also remember that the men who conquered the cold lived under 
new and rigorous climatic and dietary conditions, and that these condi- 
tions must have exerted a strong influence upon the more plastic elements 
of their bodily form. Therefore, metrical and morphological differences 
in physical type which appear, during the course of these millennia, may 
imply, in some instances, a response to environment rather than a diver- 
sity of origin. 

From all of the regions in Europe which we know to have been inhab- 
ited during the Upper Palaeolithic period, over one hundred skulls which 
have been disinterred at one time or another, during the last century, 
have been called to the attention of persons competent to determine their 
age. Of these hundred or more skulls, however, only sixty odd have been 

Hooton, E. A., AJPA, vol. 1, 1918, pp. 53-76. 


measured and published. We have, in this group, a large enough series 
to merit treatment by biometric methods, in contrast to the remains of 
earlier non-human species, which consist for the most part of single speci- 
mens, and may, therefore, be approached from a morphological and 
anatomical standpoint alone. 

Unfortunately, these crania have not been drawn in equal proportions 
from all the countries in which Upper Palaeolithic cultures are repre- 
sented. By far the largest number come from France, where they were 
preserved in caves, and where archaeological interest, over an entire cen- 
tury, has been greater tnan in any other European country. A smaller 
number come from England, Spain, northern Italy, Germany, Czecho- 
slovakia, Poland, and Russia. In studying this group of skulls as a whole, 
we must remember that the western European element is over- weighted. 

Morant, the present leader of the English biometric school, has con- 
tributed a valuable statistical study of these skulls. 21 (See Appendix I, 
col. 1.) To twenty-seven, which he personally remeasured, he adds 
twenty-five measured by other investigators. These fifty-two skulls, of 
unquestioned geological age, form the nucleus of his study. Of these 
skulls, seventy per cent come from the first, or Aurignacian period. 

Although in the later sections of this chapter we shall examine the posi- 
tion of these skulls singly by regions and by periods of time, it will be 
profitable, for the moment, to follow Morant in treating this group of 
crania as a single unit. Despite the fact that the Europeans of the Upper 
Palaeolithic were probably the product of more than one invasion, and 
despite the fact that they lived through a long period of time, and covered 
a geographical range which includes the greater part of the continent, 
the first of several striking results of Morant's study is the discovery that 
this composite sample is little more variable in the totality of its features 
than one would find in any large cranial collection of post-glacial men, 
unified in space and in time. Von Bonin, working with the long bones 
and extremities, obtained exactly the same result. 22 

It is amazing to find that the Upper Palaeolithic men were less variable, 
on the whole, than the inhabitants of London who were buried in plague 
pits during the seventeenth century. They were less variable than the 
modern rural population of a small section of Carinthia, and only a little 
more so than the skulls of the extremely isolated Greenland Eskimo, 
whose time span covered at most a few centuries, or the Egyptians who 
were buried at Gizeh between the twenty-sixth and thirtieth dynasties. 

The great complexity of race in modern Europe fs largely due to post- 
Pleistocene migrations from other continents, and the retention of local 

81 Morant, G. M., AE, vol. 4, 1930-31, pp. 109-214. 
22 Bonin, G., von, HB, vol. 7, 1935, pp. 196-221. 


types in modern populations reflects the greater isolability in small regions 
of farmers than of hunters. But the Upper Pleistocene people were by no 
means completely homogeneous, as will be shown later by an examination 
of individual crania, in their chronological and geographical contexts. 

Since, as Morant has shown, this total Upper Palaeolithic group is 
unified enough to be considered a single population, 28 we may proceed to 
generalize about the traits which most of the members of this group pos- 
sess in common. The first and most notable of these is the extremely large 
size of the brain case, larger in most cases than Galley Hill or most modern 
men, and comparable in size to Skhul. This is found in all but a few of the 
skulls, whatever the actual dimensions and forms. The cranial indices, 
however, are very variable, ranging from sixty-five to eighty-five, and this 
variability is too great to imply a single homogeneous type. 

In these skulls the males are easily distinguished from the females, for 
there is a greater difference between the sexes than is usual among more 
recent groups of man. The same is true of long bones and stature. 24 This 
implies, of course, a stronger development of secondary sexual charac- 
teristics. In the male skulls the bony markings are all pronounced, the 
browridges are as a rule heavy, the faces are excessively broad, with flar- 
ing zygomata. The upper face height is variable medium to short in 
most individuals, but in others quite long. 

One of the most distinctive characteristics of most (but not all) of these 
skulls is that the orbits are very broad and very low. The nasal skeleton 
is almost always prominent. 25 The nasal root, although deeply overhung 
by glabella, is still high, and the osseous nasal profile is as a rule straight 
or convex. The nasal spine is sharp and the lower border well marked. 
The nose, on the whole, is leptorrhine to mesorrhine. 

The lower jaw presents just as marked an individuality as does the 
cranium. This bone is deep, wide, and heavy, with flaring gonial angles 
and a prominent chin. The palate is rather wider than those of most 
living men, although the teeth are not<of excessive size. If one judges the 
face form from the calvarium alone, the great breadth of the face, coupled 
with a variable length, yields in most cases a low upper facial index, plac- 
ing these skulls in the euryene category. If, however, one calculates a 
total facial index, many of these skulls are leptoprospic, for the great 
height of the mandibular symphysis compensates for the shortness from 
nasion to alveolar point. This condition, in which the lower part of the 

23 By the word population we do not, necessarily, mean a human aggregation of 
single racial origin. What we mean here is a group of people, unified by interbreeding 
and forming a geographical and social unit. Such a population, of course, may have 
a multiple origin. 

24 Bonin, G., von, op. cit. 

26 The Grimaldi "negroids" form an exception. 


face is exaggerated, is one of the chief diagnostic features of this type of 
man, and a suggestion of it may still be seen among some of the living 
peoples of northern Europe. 

In the totality of facial features, with a few exceptions, the Upper Pa- 
laeolithic people may be said to have resembled modern white men. 
Some, however, probably looked like a certain type of American Indian, 
notably that of the North American Plains, and of the OnaS and Tehuelche 
of southernmost South America. This comparison, we must remember, 
is wholly morphological, since we do not know Upper Palaeolithic man's 
pigmentation, hair form, or hair distribution. 

The skeletons of the Upper Palaeolithic people vary in size by sub- 
periods, as will be shortly demonstrated, but as a whole the group was 
tall, long-limbed, and slender, with narrow hips, broad shoulders, and 
large hands and feet. On the whole, the limb bones were not excessively 
robust, and the limb ratios, determining the relative lengths of arm and 
leg segments, and of arms to legs, were unstable. 

The mean stature of the males was about 173 cm., of the females 1 55 cm. 
The men were taller than the means of any modern European countries, 
with the exception of Iceland and Montenegro, but not taller than mod- 
ern Americans. The women, on the other hand, were actually small. 
The equivalents of these mean statures are, in feet and inches, but five 
feet nine, and five feet one and a half. Galley Hill man, by comparison, 
was only five feet two. 

Morant, in his statistical study, compared his Upper Palaeolithic sample 
with a long list of post-Pleistocene cranial series. He found that the early 
group exceeded all of the later ones by a wide margin in seven measure- 
ments, 28 while it reached the limit of recent human means in six others. 27 
This mass deviation would, in Morant's opinion, place Upper Palaeo- 
lithic European man at one end of the scale and the rest of humanity, 
white and otherwise, all of lesser antiquity, at the other. 

It is possible to quibble with Morant, and to discover small series or 
subseries which contradict this finding. For example, the Ona skulls from 
Tierra del Fuego, 28 a series of Bronze Age crania from Esthonia, 29 and of 

M Horizontal circumference, glabcllooccipital length, sagittal arc from glabella to 
opisthion, nasio-bregmatic arc, internal biorbital diameter, bizygomatic diameter, and 
length of the foramen magnum. 

** Transverse circumference, brcgma-lambda arc, biasterionic breadth, bimaxillary 
breadth, orbital breadth. These five, according to Morant, fall within 1 mm. of the 
greatest post-Pleistocene means. Orbital height, he finds, is .7 mm. shorter than the 
lowest comparative mean. 

28 Morant, in his group "Fuegians, pooled," mixed Ona skulls with those of the 
smaller and quite different Yaghans. If one abstracts the Ona crania from Lebzeltcr's 
original tables, he will find that the European Upper Palaeolithic means of Morant are 
essentially duplicated. w Friedenthal, A., ZFE, vol. 63, 1931, pp. 1-39. 


Iron Age ones from the Norwegian coast, 80 are equally large in facial as 
well as cranial dimensions. But these exceptions in no way invalidate his 
discovery, that the Upper Palaeolithic people, despite their generalized 
European facial appearance, were separate in many metrical characters 
from most of living, or for that matter pre-Aurignacian, sapiens men. The 
reason for this deviation is not difficult to discover, but we must approach 
the obvious conclusion slowly, in order to make sure of an accurate recon- 
struction of prehistoric events. 


The Aurignacian flake culture, 31 with which the Upper Palaeolithic 
period in Europe began, was not a single unit throughout its time span, 
but seems to have been composed of several separate entities derived from 
more than one non-European source. 

The first Aurignacian level in Europe, the Chatelperronian, is repre- 
sented by three skeletons only. These include the two "negroids" from 
the Grotte dcs Enfants, Grimaldi, near Mentone, and Combe Capelle. 
Of the three, the Grimaldi pair may have been the older. Except that 
they belonged to the earliest Aurignacian period, a more exact estimate 
of their age is not possible. 32 These were the remains of an adult female 
and an adolescent male. Disregarding for the moment their racial affin- 
ities, we may be sure that they were fully sapiens, and that they resembled 
Galley Hill in stature and in gross cranial vault form. The vault dimen- 
sions, however, are smaller. They thus show nothing whatever of the 
great size and robusticity of the crania belonging to the total Upper 
Palaeolithic group, and nothing of the latter's exuberance of bodily 

In other respects they were apparently somewhat negroid, in the sense 
that they possessed features divergent from the modern white standard 
in a modern negroid direction. 38 These include the virtual absence of 
browridges, a sharp bowing of the frontal bone, low, broad nasal bones, 
a guttered lower border of the nasal opening, alveolar prognathism, a 
large palate, and large teeth. The orbits, furthermore, are relatively 
narrow, the face both absolutely short and narrow. The long bones show 
a difference, however, in limb proportions between these people and 

wSchreiner, K. E., SNVO, II, #11, 1927, pp. 1-32. 

Garrod, Miss D. A. E., RBAA, Pres. Ad., Sect. H., 1936, pp. 155-172. 

M Hopwood, A. T., op. cit. 

M Verneau, R., "Anthropologie," in Les GrotUs de Grimaldi, vol. 2, fasc. 1. 

Keith (Antiquity of Man, p. 67) and Morant (AE, vol. 4, 1930, pp. 116-119) deny this 
negroid character completely, Morant on the grounds of faulty restoration, which exag- 
gerated the prognathism of the adolescent male. 


Galley Hill, for the distal segments are relatively long, and the arms long 
in relation to the legs. 34 In this as in the possession of long heels 36 the 
Grimaldi specimens are truly negroid, and again upset the unity of the 
total Upper Palaeolithic sample. 

There is no type of man more completely sapiens 36 than a negro. The 
two Grimaldi specimens, in being partially negroid variants or relatives 
of the Galley Hill group, are entirely divorced from the line or evolution 
which produced either the Palestinian Skhul people or the later European 
Aurignacians. In this respect the argument as to how much or how little 
negroid they actually were, is of no importance. 

Whence they came to Europe, in the van of the Upper Palaeolithic 
migrations, is likewise not, at the moment, worthy of extensive argument. 
They must have come from Africa or southwestern Asia, but until others 
of the same type have been found, the problem will remain open. They 
may represent an early negro-white mixture, or a generalized proto- 
negroid in the process of specialization. 37 They are probably too late, 
however, in time, to have been contemporary members of the generalized 
stock which may have been mutually ancestral to the negroes and whites. 

The study of the third specimen, the male skeleton of Combe Capelle, 
is more pertinent to our present problem. Like the two Grimaldi negroids, 
it deviates completely from the body of Upper Palaeolithic crania and 
long bones in the distinctive features of the main group. Although as 
long as the mean for the total series, the cranial vault is considerably 
narrower, somewhat higher, and smaller in capacity. In the details of 
vault form, it is essentially similar to Galley Hill, but is actually narrower 
even than Galley Hill itself. 

The Combe Capelle face is the earliest which can be definitely asso- 
ciated with a Galley Hill type of vault. Here it differs again from the 
middle Aurignacians to follow, for the bizygomatic and bigonial diame- 
ters are as small as those of most modern long-headed white men, and the 
face, orbits, and mandible are narrow. The nasal opening is wider than 
those of most later Upper Palaeolithic European crania but the nasal bones 
are European in form. There is no prognathisrn and the subnasal seg- 
ment of the face is not exceptionally large. Like Galley Hill, Combe 
Capelle man was short, with a stature of 160-162 cm. 

It will be necessary, in studying the crania of the remainder of the 
Aurignacian, to combine all sub-periods, since no distinction has been 
made in the majority of cases. Furthermore, the crania which might 

34 Bonin, G., von, op. cit. 9 p. 205. 

//., p. 215. 

38 As opposed to Neanderthaloid. 

Montandon, G., RA, vols. 4-6, 1936, pp. 105-139. 



have been Solutrean are also included. The present survey, then, in- 
cludes the famous Cr6-Magnons of France, and the Moravian mammoth 
hunters, who lived in the open and buried their dead in sepulchres built 
of mammoth jaws and shoulder blades. 

Despite the general homogeneity of Upper Palaeolithic man, these two 
groups, the western and eastern, may be shown to have differed from each 
other in certain well-defined ways. Both were tall, with statures well 




Reconstructions, under the direction of Professor V. Lebzelter; Fig. 13 by Herr 
Fahrwickel, Fig. 14 by Herr E. Grenzer. MAGW, vol. 65, pp. [4], [26]. 

over 170 cm., and in this likeness of growth they segregate themselves 
from the few early Aurignacian representatives which we have studied. 

Both the western and the eastern type possess the special characteristics 
of Upper Palaeolithic man which have been described earlier. But there 
is one principal feature which .separates them the cranial index. The 
Cr6~Magnons, who were concentrated in France, range in this ratio from 
69 to 85, whereas the eastern group, including the Russian skulls, ranges 
from 64 to 76. The mean index for the French skulls equals 76, while that 
of the eastern group, representing the same period of time, is 71. In other 
words, the eastern Aurignacian type, like Galley Hill and Combe Capelle, 
was purely dolichocephalic, while the Middle and Late Aurignacians of 
France include among their numbers a brachycephalic element, which 
reaches the high limit of 85 in the male skull, Solutr6 #2. 


The western group has been named Cro-Magnon after a senile male 
skull which is .usually taken as the standard example of this type, and 
which possesses the most specialized Upper Palaeolithic features in an 
exaggerated manner. Skulls of this type have a rather flat vault, the 
lowest and broadest orbits in the entire series, and short, extremely wide 
faces. Their nasal apertures are of medium width, and their nasal bones 
highly curved and projecting. 

The most brachycephalic skull shows, of course, quite a different con- 
formation of the cranial vault. It belongs to the curvoccipital type, with 
a gently rounded rear profile. This Solutr #2 specimen is a large skull, 
and belonged to a tall man. Its face, however, shows the typical Cr6- 
Magnon features of flaring zygomata and wide jaw, combined with 
extremely low orbits. The Cr6-Magnon character of this face, while 
marked, is not as clearly shown as in the dolichocephalic examples, for the 
fuller bulge of the temporals obscures it. The Cr6-Magnon type, in the 
widest sense, therefore, includes both long-headed and round-headed ex- 
amples with transitions in between, and the features which differentiate 
it are just as pronounced in the roundest skulls as in the more numerous 
narrower ones. 

Let us turn to central and eastern Europe, and study the purely long- 
headed examples from this part of the continent. In general, they re- 
semble an exaggerated and leptorrhine Combe Capelle, with the low 
orbits, wide faces, and heavy jaws found in excess further west. A few 
skulls deviate in various ways from the standard, however; of these three 
are notable, Briinn #1, Lautsch, and Predmost #3. 

Briinn #1 , which lacks a face, but possesses a mandible, is, in vault form 
and size, and in the lower jaw, the duplicate of Combe Capelle. Lautsch, 
which has a face, belongs partly to the same general class, but is broader. 
Its face, however, is narrow, and in this conforms to the Combe Capelle 
type. On the whole, the eastern skulls, while subjected to the same in- 
fluence which brought about an increase in gross size and osseous extrav- 
agance in the Cr6-Magnons, nevertheless cling closer to the older Galley 
Hill form, and were not affected by whatever factor caused the brachy- 
cephaly in some of the western specimens. 

The third of the not fully typical eastern crania, Predmost #3, 88 is of 
great value, for it reveals in a certain manner the reason for the 
general- peculiarities of the Upper Palaeolithic series as a whole, and for 
their separation, shown by Morant, from the bulk of living humanity. 
This reason is simply that P?edmost #3 resembles Skhul #5 very closely, 
both morphologically and metrically, while neither of these two speci- 
mens deviates notably from the Upper Palaeolithic metrical means. 

88 This may even have been associated with a Solutrean culture. 


On the whole, the Upper Palaeolithic group, including Predmost #3, 
is intermediate between the Galley Hill-Combe Capelle type and the 
Neanderthals, as known to us from the European Neanderthaloid group. 39 
In the first place, the horizontal circumference, taken above the brow- 
ridges, ranges from 538 to 563 mm. in male Neanderthals. The Upper 
Palaeolithic means is 549.1 mm., the individual figure of Pfedmost #3 is 
556, that of Combe Capelle, 527 mm., which would be nearer a modern 
dolichocephalic mean. In face breadth, the Neanderthal figure is repre- 
sented by La Chapelle aux Saints with 152 mm., and the Le Moustier 
adolescent with 148 mm. The Upper Palaeolithic mean is 142.8 mm., 
Pfedmost #3 is 144 mm., and Combe Capelle 137 mm. Again, Combe 
Capelle represents modern European man, and the Upper Palaeolithic 
group takes an intermediate position. 

The same intermediate position is found in a number of other charac- 
ters, including the vault breadth and height, the minimum frontal diam- 
eter, the widths of the orbits, and the distance between the orbits. In in- 
dividual cases, such as Pfedmost #3, the upper face height is intermediate 
also, but in the group as a whole it is not, for the shorter dimension pre- 
vails. The same is true of the nasal dimensions in which Upper Palaeo- 
lithic man is not perceptibly Neanderthaloid. The cranial lengths of the 
Upper Palaeolithic group are no greater than those of Combe Capelle and 
Galley Hill; in fact, frequently shorter. The reason for this may be that the 
equivalent Neanderthaloid diameter includes the browridges, which, when 
eliminated, make the brain length somewhat less than that of Galley Hill. 

The stature of the Upper Palaeolithic group equals that of Skhul; the 
sex differentiation in size is the same; the pelves are similar, and so is 
the rib section. The hands and feet are likewise large. 

If the European group, with the exception of P?edmost #3, is less 
Neanderthaloid looking than Skhul, this is not surprising. The distance 
in time between the two was probably as great as or greater than that 
between the beginning of the Middle Aurignacian and the present. Fur- 
thermore, we have been comparing European Upper Palaeolithic skele- 
tons with those of European Neanderthals. While the Aurignacian hunters 
of Europe may possibly have absorbed some of the local Neanderthal 
survivors, it is likely that the main accretion of this element took place 
farther east, 40 and we do not know that those so accreted were as spe- 
cialized in a non-sapiens direction as the European examples. 

89 Measurements from Morant's tables in AE, vol. 2, 1927, pp, 376-377. 

40 The Aurignacian culture does not seem to have absorbed the local Mousterian of 
western Europe, except in a few doubtful instances. See Leakey, L. S. B., Stone Age 
Africa t pp. 185-186. On the whole, the theory that European Neanderthals had become 
extinct, or had departed before the arrival of the Aurignacians, is not invalidated by the 
discovery of the Neanderthaloid characters of Middle and Late Aurignacian man* 


In admitting the partially Neanderthaloid character of Upper Palaeo- 
lithic man (which is no new theory), we must accept at the same time 
some genetic principles which apply to modern primary crosses between 
distant races, 41 as well as to these ancient interspecific mixtures. Although 
blending is the rule in most characters, simple dominance appears in a 
few; while major changes in size appear through this mixing. The stature 
of the hybrids far exceeds that of their parents, and through the general 
genetic upset, the brain size becomes greater than that of the earlier purely 
sapiens stock. 

It must be admitted that there is an alternative interpretation of the 
Neanderthaloid traits of Upper Palaeolithic man. That is that he repre- 
sents a stage in the evolution of Neanderthal in a sapiens direction; that 
different branches of the Neanderthaloid stock evolved into sapiens men 
at different times; and that the Swanscombe Galley Hill Kanam 
Kanjera stock went through this process at a much earlier date than 
did the group under consideration. 

For the purposes of the present study, it makes no difference whether 
the early sapiens stock, fully evolved by the Mid-Pleistocene, had passed 
through a Neanderthaloid stage in its previous history, or had evolved 
directly from some less gerontomorphic and more gibbonoid ancestor. 
The question which is at the moment pertinent is, are the Skhul Upper 
Palaeolithic peoples to be considered Neanderthaloid-^aj&zVn^ hybrids, or 
simply evolved Neanderthaloids, in which case the hybridization con- 
necting the Upper Palaeolithic people with modern Europeans would 
have occurred later? All of the existing evidence, of somatology as of 
archaeology, points to the former hypothesis, which we have accepted, in 
lieu of further information, as one of the main theses in our reconstruction 
of European racial history. 

Returning to the specific consideration of the Upper Palaeolithic Euro- 
pean group, we find that the difference between the eastern and the 
western Aurignacians, which consists, most demonstrably, of the brachy- 
cephalic tendency in the latter, has not been explained. It might, how- 
ever, have been due to a differential mixture between sapiens and more 
than one Neanderthaloid strain. The Neanderthaloids in Europe, who 
lived in the western part of the continent, varied in cranial index from 67 
(Gibraltar) to 77 (La Quina), and were not far from the French Upper 
Palaeolithic mean of 76. 

When measured from ophyron, a point on the frontal bone behind the 
browridges, the crania of these Neanderthals have the following lengths: 
three males, 193, 186, 187; three females, 185, 183, 186 mm. These are 

"Shapiro, H. L., MBM, vol. 11, 1929, pp. 1-106; The Heritage of the Bounty, 
pp. 217-233. 


shorter than the French Upper Palaeolithic means, taken from the same 
point, of 195.6 mm. for males, and 188.6 mm. for females. The cranial 
indices calculated from these lengths are, in five out of eight Neanderthal 
cases, above 80.0. Thus there was, in the Neanderthal group as we know 
it, a brachyencranial, or brachycerebral, tendency in brain form which, 
with a reduction of browridges, might, in mixture, have caused brachy- 
cephaly in some of the hybrids. That it may have done so in the case of the 
French brachycephals, notably Solutre* #2, which has a cranial length of 
182.5 mm., is by no means more than a suggestion. 

As the reader will have gathered from the preceding pages, the study of 
race in Europe during the advance of the last ice is not a simple matter, nor 
one to be solved lightly. It will be of help to study parallel developments 
in other quarters, especially in Africa. 


During the Late Pleistocene, at the time of the Wurm glaciations in 
Europe, northern Africa, including the present Morocco, Algeria, and 
Tunisia, enjoyed a cool climate and an abundant plant life; making an 
admirable home for human beings. Fortunately, many Late Pleistocene 
skeletons from these countries have been studied, and we are able to sup- 
plement our information from Europe very greatly by comparison. 

During the Upper Palaeolithic, there were three cultures in North Africa 
which existed contemporaneously as geographical units; the Capsian, 
covering a restricted range in Tunisia and eastern Algeria; the Oranian, 
a related culture extending over the provinces of Alger and Constantine, 
and into Morocco; and the Aterian, along the Moroccan seaboard. 

The Capsian and Oranian were cultures basically related to the Aurig- 
nacian of Europe, but which contained throughout their history a micro- 
lithic blade element which was destined to move northward, after the end 
of the glacial period, and to invade Europe as the Tardenoisian. The Cap- 
sian had probably moved westward along the southern Mediterranean 
shore from the East; the Oranian was nothing but a western extension of 
the Capsian; while the Aterian was a protracted survivor of the Mouste- 
rian, which developed its own peculiarities as time went on, and was grad- 
ually crowded to the Atlantic seaboard by the Oranian. 

It is at present believed that North Africa, during the Late Pleistocene, 
was a marginal area of refuge, and not a highway of cultures. Gibraltar 
served less as a bridge than as a barrier. Nothing can better attest the 
passive cultural r61e of North Africa during this period than the fact that 
the Aterian, a Mid-Pleistocene culture, was allowed to elaborate its own 
special technique long after the Mousterian, from which it sprang, had 
passed out of existence elsewhere. 


That the earlier phases of the Capsian and Oranian, coming directly 
after the Mousterian, were comparable in time to the Upper Palaeolithic 
cultures of Europe, such experts as Menghin, Obermaier, and Leakey are 
unanimous, 42 while Miss Garrod, on the basis of Vaufrey's work, would 
make them later. 43 

We can only, at this point, agree with Menghin that while the exact 
time correlation of North African and western European Late Pleistocene 
industries is still floating, they may be considered roughly parallel. At 
present, the general agreement is that the essential elements of both 
European and North African Upper Palaeolithic cultures came from the 
east, and had, at least in part, a common origin. 

So far, all of the human remains of the Late Pleistocene from North 
Africa come from the Province of Gonstantine, where most of the archaeo- 
logical work has been done. The total of these skeletons probably reaches 
one hundred, but, unfortunately, less than half of them have been fully 
preserved or competently studied. 44 They come for the most part from two 
great sites, Afalou bou Rummel, and Mechta el 'Arbi. The former is 
Oranian, the latter Capsian. 

Afalou bou Rummel is an Early Oranian site. Within this early horizon, 
Arambourg distinguishes two levels, a lower^and an upper. The lower 
may be correlated with the Early Aurignacian of Europe, by one system, 
or with Middle and Late Aurignacian, according to another. 

The lower level is represented by the skeleton of a single adult male, to 
whom we shall refer by his catalogue number, #28. Number 28 was a 
short man, about 161.5 cm. tall, equivalent in stature to Galley Hill, 
Combe Capelle, and the male negroid from Grimaldi. His skull differs 
greatly from the others taken from the upper level of the same site. It is 
ovoid in shape, hyperdolichocephalic, and low vaulted; it possesses a slop- 
ing forehead, a large U-shaped palate, and high orbits. It is only moder- 
ately massive, and is about equal in this respect to Combe Capelle. This 
skull is that of a generalized white type, and can be placed without much 
difficulty into the general class of Galley Hill and Combe Capelle. Like 
the latter, its nasal aperture is wide, its index chamaerrhine. 

Forty-nine other crania have been taken from the upper lenses of the 
Early Oranian culture level at Afalou bou Rummel. These correspond 

42 Menghin, O., Weltgeschichte der Stein&it, pp. 34-35. 

Obermaier, H., AAnz, vol. 7, 1931, pp. 259-265. 

Leakey, L. S. B., Stone Age Africa, pp. 105-111. 

48 Garrod, Miss D. A. E., RBAA, Pres. Ad., Sect. H. 

Vaufrey, R., Anth, vol. 43, 1933, pp. 457-483. 

44 But two satisfactory accounts have been published : 

Part II of Les Grottes Palaeolithiques de Bern Seghoual, by Boule, Vallois, and Verneau. 

Cole, Fay-Cooper, LMB #1, 1928, Section on skeletal material. 


closely in physical type to Middle and Late Aurignacian man in western 
Europe, but the two groups are not identical. Like Cr6-Magnon, all of the 
Afalou skeletons studied were tall, with an estimated male mean falling 
between 171 and 175 cm., according to different methods. Their limb 
proportions, with long distal segments, are like those of many of the 
Cro-Magnon group; while their hands and feet, similarly, are both longer 
and broader than those of most Europeans. The combined height of the 
vertebrae show that their bodies, as well as their legs, were long, and 
the total bulk of a typical male, in good condition, must have been great, 

A high ratio between the length of the collarbone and that of the upper 
arm (clavico-humeral index) reveals that they had broader shoulders than 
those of most modern white men, a feature which has been also noticed 
on the Chancelade and Obercassel skeletons, and perhaps is equally true of 
the European group as a whole. The pelves are high and have narrow 
openings; the feet are highly arched, with well-developed heels; and the 
size and muscular markings of the long bones differentiate the males 
from the females clearly. All of the bodily traits of these men are shared 
by Cr6-Magnon, and all are, in a general sense, European. 

The Afalou crania have been exhaustively described and thoroughly 
illustrated. In general, they are very large, low-orbitted skulls, thick-boned, 
and marked in high relief for muscular attachment. The browridges form 
a heavy jut, even greater in most instances than those of the Gr6-Magnons. 
Behind a salient glabella the forehead slopes in all instances. Vertical 
foreheads, frequent among modern whites, especially females, and present 
in some Cr6-Magnon individuals, do not occur here. The union of the 
parietal and occipital bones is always marked by a lambdoidal depression, 
or flattening, 46 while below this depression the occiput is usually bun- 
shaped and projecting. The mastoids are strongly developed, and the 
thickness of the vault is greater than that of modern man, but no greater 
than with Gr6-Magnon. 

Metrically, the male skulls (see Appendix I, col. 3) are practically 
identical with those of the 'total European series, except that they are 
slightly shorter and higher in vault dimensions, while the upper face is a 
little shorter. In these divergences from the total European group, they 
resemble the western branch, or Cr6-Magnons. The cranial indices of 23 
males range from 70 to 80, with a mean of 74.8; while the female figures 
are: range 70 to 84; and mean 75.7. Both in range and in means of 
head form, the Afalou series equals that of Cr6-Magnon. 

The nose of the Afalou type is perfectly European in bony conformation. 

45 This feature is extremely common among living North African tribesmen, and 
among crania from the Canary Islands. See Coon, C. S., Tribes of the Rif, p. 312. 
Hooton, E. A., The Ancient Inhabitants of the Canary Islands, p. 134. 


The paired nasal bones unite at a sharp angle, without trace of flatten- 
ing, while the bridges are high and mostly convex. The nasal spine is 
strong and projects far forward. The nasal index, which lies just over the 
border of chamaerrhiny, 46 furnishes a real metrical difference between 
Afalou and Cr6-Magnon. The elevation of the index is due to a shorter 
height as well as to a greater width. Not one of the Afalou skulls is actually 
leptorrhine. This feature, combined with the sloping forehead and heavy 
browridges, serves to differentiate the types in the two continents. The 
Afalou mandible, furthermore, is extremely broad, deep, and heavy. In 
the possession of a pronounced chin greater than those commonly found 
among the living, it is clearly opposed to any known Neanderthal form. 
However, it resembles the Neanderthaloids in one feature; the bigonial 
breadth is frequently greater than the mandibular length, a condition 
rare in Homo sapiens, and not even found in Skhul. 

In the Cr6-Magnon series, the combination of a short, broad upper face 
with a long cranial vault has often been called "disharmonic," and it has 
been asserted that this condition is the result of mixture between a longer, 
narrower-faced dolichocephal and a shorter, wider-faced brachycephal. 47 
In the European series, although both long and round skull forms occur, 
there are not enough crania which still possess facial bones to make a 
statistical analysis of this point valid. But in the Afalou series, where the 
same set of conditions is duplicated, such an analysis is possible. 48 Out of 
nine dolichocranial skulls, four have upper facial indices in the broad cate- 
gory, while fourteen out of eighteen of the rounder-headed examples are 
broad faced. The tendency toward a broad upper face form, then, is 
borne predominantly by the meso- and brachycranial element in the group. 
If this is true for the Afalou series, it is probably equally valid in the Cr6- 
Magnon group. 

Two of the Afalou skulls, however, present the "disharmonic" combina- 
tion of a hypereuryene face with a long skull form. In explaining this 
anomaly we must remember that an extreme width of the face is sex- 
linked in both the Cr6-Magnon and Afalou series; it is a manifestation of 
the extreme ruggedness and luxuriance of muscularity which the males of 
both series manifest, and is lacking, as a rule, among the females. 

One other peculiarity which is common to both the European and the 
North African Upper Palaeolithic peoples is a very low orbital index. This 
again lends itself in the second series to statistical analysis. Only three out 
of eleven dolichocranial skulls are chamaeconch, while fourteen out of 


Mean for 21 males 53.1; Nose Ht. = 52.7 mm.; Nose Br.^28.4 mm. 

47 Hooton, E. A., Canary Islands, pp. 204-207. 

48 The coefficient of mean square contingency between the cranial index and the up- 
per facial index,, calculated with nine boxes, equals .53. 


eighteen of those with higher cranial indices fall into this low-or bitted 
category. 49 

Hence we may deduce that the two parallel series, Cro-Magnon and 
Afalou, consist in each case of a Galley Hill-Neanderthal mixture as a base, 
with which is associated a variant tendency to round-headedness. To this 
is linked an extremely short, broad upper facial form, with a heavy lower 
jaw, and wide, low orbits. At the same time, certain differences, such as 
the nose form, definitely prevent the assumption that the two are identical, 
and make it extremely unlikely that the two met, after their initial separa- 
tion, during the entire span of the Late Pleistocene. 

Other facts strengthen this conclusion. 50 The Afalou people knocked 
out one to four incisor teeth from the jaws of each person of either sex, be- 
tween the ages of fourteen and sixteen, apparently as a puberty rite. Tooth 
knocking is unknown in Europe before the Mesolithic, 51 although finger- 
chopping, during the Upper Palaeolithic, is indicated by the outlines of 
mutilated hands on the walls of the caves. Therefore, if the Cr6-Magnon 
people observed bloody puberty ceremonies, as is quite possible, they must 
have removed some less tell-tale part of the anatomy than the teeth. While 
this bit of cultural evidence renders the theory of physical contact between 
the two groups unlikely, it does not necessarily affect the problem of rela- 
tive age. We still do not know whether the Afalou men, whose sequence of 
types parallels that in western Europe, were contemporaneous with their 
kinsmen to the north, or later than them to arrive. 

From a study of these presumably Pleistocene Algerians, we are able to 
confirm the conclusions reached in the preceding section, and to amplify 
them. A fully sapiens individual, comparable to Combe Capelle in every 
important respect, preceded, in time, a group of overgrown, large-headed 
and wide-faced Neanderthal-sapiens hybrids. This latter type, like Cr6- 
Magnon and unlike the people in central and eastern Europe, bore a 
tendency to brachycephaly. That Cr6-Magnon and Afalou men were the 
parallel termini of similar movements, and not way stations on a single 
line of migration, is probable. In view of the earlier evidence of a similar 
mixture in Palestine, and of the general center of Aurignacian activity in 
that neighborhood, it may be considered likely that the second pair of 
parallel movements proceeded westward from that general quarter. The 
earlier waves which brought Combe Capelle and Afalou #28 must have 

49 C sa .47. In this contingency table, of 6 boxes, the progression is constant, and of 
undoubted significance. 

50 It was at one time thought that the presence of caries in a small percentage in the 
Afalou teeth made them later in age than the Europeans. However, the two Solutrean or 
Magdalenian skulls from Le Roc, Charente, were also carious. 

Boule, M., and Vallois, H., BIPH, #18, 1937. 
61 See Chapter III, p. 68, footnote 27 (Ofnet). 


come from a different center. Whether the Cr6~Magnon and Afalou 
people derived their brachycephalic tendencies from parallel mixtures at 
the terminal points of invasion or brought them with them in the first 
place, cannot be determined without further evidence. 


The Aurignacian culture, the racial connotations of which we have just 
reviewed in Europe, Palestine, and, in the guise of Capsian and Oranian, 
in North Africa, also extended southward to East Africa. 62 Here, as in 
Algeria, microliths were present in the midst of other forms more typical of 
Europe and western Asia. 53 While the correlation of the pluvial periods 
of East Africa with the glacial phases farther north is still under discussion, 
it is probably safe to conclude that the Kenya and Tanganyika Lower 
Aurignacian was roughly equivalent in time to Upper Palaeolithic hori- 
zons elsewhere, although the Upper Aurignacian may have lasted much 
later, in view of the fact that East Africa was a racial and cultural frontier 
and a marginal area. 

The remains of six Upper Aurignacian men have been discovered in the 
two colonies named. Five of these were exhumed by Leakey at Gamble's 
Cave, Elementitia, B4 and the sixth is the famous Oldoway skull discovered 
by Reck in 1914. 56 Two of the Gamble's Cave specimens, and Oldoway, 
which are all masculine, consist of nearly complete skulls and long bones. 
The others from Gamble's Cave are too fragmentary to be of much value. 

In general, these specimens belong in the purely sapiens category, as 
represented by Galley Hill, Kanjera, Grimaldi, Combe Capelle, and 
Afalou #28. At the same time, however, they differ from all named in one 
important respect they are extremely tall, with statures of 177, 179, and 
180 cm., which even exceeds the Cr6-Magnon and later Afalou figures, but 
the great stature is unaccompanied by the broad shoulders and bodily 
bulk of the hybrid Europeans and North Africans. The long bones are 
very slender, and the hands and feet small and narrow. 

The same principle of attenuation applies to the faces. In all of them, 
and especially in Oldoway, the faces are extremely narrow, and very long, 
especially in the upper segments. The browridges are weak, the zygomatic 
arches feebly developed, the mandibles light and slender, with narrow bi- 
gonial diameters, and weak, although positive, chins. The orbits are high 
and narrow, and the noses likewise. The Gamble's Cave skulls are lep- 

68 Leakey, L. S. B., Stone Age Africa, pp. 38-74. 

Garrod, Miss D. A. E., RBAA, Pres. Ad., Sect. H. 

M Leakey, L. S. B., The Stone Age Races of Kenya, pp. 47-56; Stone Age Africa, p. 172. 

56 Reck, H., Oldoway, die Schlucht des Urmenschcn, Leipzig, 1933. 

Mollison, T., and Gieseler, W., VGPA, vol. 3, 1929, pp. 50-59, 60-67. 

Boule, M., and Vallois, H., UHommejomle d*A$$dar, AIPH, Mem. 9, pp. 60-64. 


torrhine, leptene, and leptoprosopic; Oldoway is mesorrhine, and hyper- 
leptoprosopic. The two Gamble's Gave skulls are orthognathous, but 
Oldoway possesses considerable alveolar prognathism. 

In vault size, these crania resemble Combe Gapelle and Afalou #28, 
rather than the European and North African crania of later Aurignacian 
and Oranian date. Oldoway and Gamble's Cave #4 are higher and nar- 
rower than the European Upper Palaeolithic mean; Gamble's Cave #5, 
which is the skull of an adolescent, is shorter, higher, and nearly as broad. 
The foreheads are gently sloping and rounded; the occiputs projecting, but 
without the lambdoidal flattening which characterizes the European 
crania. The total impression is one of thinness and delicacy. 

In the morphology of the head and face, these three specimens are not 
exactly alike. Gamble's Cave #5, which has a cranial index of 74, is nearest 
to the European standard; while the two others, Gamble's #4, with an 
index of 71, and Oldoway, with 64.5, are decreasingly so. But they are 
closer in many ways to modern European racial types than are the Upper 
Palaeolithic skulls. They seem, however, to have been subjected to some 
influence which has made all extremities, including both limbs and face, 
extremely long and thin. One may compare this with the modern changes 
in the English stock settled in Queensland. 66 

Both of the Gamble's Cave skulls seem to be fully or nearly "white" in 
the skeletal sense, but Oldoway is, in a way difficult to analyze, per- 
ceptibly negroid. Many modern tribes of East Africa, including the 
Somalis and Masai, and the upper classes of others such as the Bahimas, 
show today the same general features which are found in these pluvial 
period skulls, particularly in Oldoway. These modern Hamites have long 
spindly legs, thin hands, and narrow wrists, while their bodies are cor- 
respondingly thin and attenuated. Their skulls are universally long, 
smoothly contoured, and lacking in strong muscular markings. Their 
noses are narrow and often highly arched, their jaws light and narrow, 
their faces long and thin. All of these modern East African Hamites show 
a certain amount of negroid admixture, 'but their skulls are considerably 
smaller than the three from the pluvial period. 

On the basis of head size, if for no other reason, these skulls cannot be 
dismissed as intrusive burials from later periods. Mollison, who has 
studied the Oldoway skull, is convinced that it is as fully fossilized as the 
bones of the other fauna of the period to which it is now attributed. 67 

M I can find no adequate references to this phenomenon, but common observation 
attests its existence. 

* T And not of the Lower Palaeolithic horizon to which it was first ascribed, Boule, M., 
and Vallois, H., AIPH, Mem. 13, 1934, pp. 60-64. 

Mollison, T., and Gieseler, W., VGPA, vol. 3, 1929, pp. 60-67. 

Leakey, L. S. B., Stone Age Africa, pp. 172-173. 


The only reasonable conclusion is that the modern Hamite is ancient in 
East Africa, and is at least partially descended from this ancestral, purely 
sapiens, form. At the same time we may be equally sure that modern post- 
glacial Europeans of Mediterranean type did not come from this particular 
corner of Africa; that whatever the date of these specimens in years, East 
Africa was not, in Upper Palaeolithic times, the center of Mediterranean 
racial evolution. 

Neither, it would appear, was the Sahara; so far the archaeologists have 
not found evidences of the Upper Palaeolithic Capsian culture in the cen- 
tral zone of the desert itself, where there is at present a gap between the 
Levalloisian and what appears to be an early, arrow-chipping Neolithic 
in Capsian tradition. The Capsian apparently came to North Africa from 
the east, and the mid-Sahara may have served even during Pleistocene 
times as a dividing line between white and negroid humanity, just as it 
does today. At the same time it is likely that the Empty Quarter of south- 
ern Arabia even in those days functioned as a barrier between Mediter- 
raneans and Veddoids. 88 Although what are now the edges of deserts may 
have been breeding grounds of white humanity during the Pleistocene, 
the great deserts themselves have always been racial frontiers. 


In concluding our survey of human racial types and racial movements 
during the Pleistocene, it will be well to return from Africa to study that 
group of Late Palaeolithic cultures found in Europe and northern Asia, 
and collectively known as Magdalenian. 59 

The Magdalenian was the longest of the Upper Palaeolithic cultural divi- 
sions in time span, lasting in western Europe from the peak of Wurm II until 
about 11,800 B.C., while in parts of eastern Europe where it is found it may 
have been even older. It marks the culmination and decline of the second 
Wurm advance, and is the first instance, except for that of the geograph- 
ically limited Solutrean culture, in which we are sure that sapiens man 
was exposed to the full force of a glacial climate. During the Magdalenian 
as during the Solutrean, the great abundance of fine needles and other 
tailoring implements in archaeological deposits attests the close cultural 
adaptation of these people to the cold conditions under which they hunted. 

During the Magdalenian there must have been numerous population 
shifts and migrations with the changing climate, as the men followed the 

68 See Chapter XI, section 6, for a discussion of the Veddoids in southern Arabia. 

69 A separate study of race during the Solutrean has been omitted, since there are no 
skulls which all authorities accept as definitely belonging to that short and far from 
widespread cultural phase. Those of Pfedmost, including jf3, might well be Aurig- 
nacian; those of Le Roc fit more easily, from the craniological standpoint, into a Mag- 
dalenian category. 


herds of reindeer which formed their chief article of diet. We know that 
at one time reindeer crossed the Pyrenees and wandered into Spain, where 
the Magdalenian hunters followed them. In general, however, Spain, like 
Italy, was a marginal area, relatively sheltered, in which local cultures of 
Aurignacian origin persisted, with the addition of microlithic elements, 
presumably from Africa. 60 These southern inroads were, however, but 
minor Magdalenian incidents. It was a sub-glacial tundra culture, and 
stretched eastward across Siberia, where numerous sites have recently 
been discovered. 

Despite its great time expanse, the Magdalenian is represented by fewer 
skeletal remains than is the Aurignacian. The finds seem to be limited 
entirely to the west 61 to France, England, western Germany, and 
Spain. 62 Moving eastward from Germany, we find no more human re- 
mains until we arrive in northern China. 

The number of fully authenticated Magdalenian skulls, about twenty- 
five, might be large enough to warrant separate statistical study if all of 
them were well documented. As matters stand, we are able to discuss but 
a few of them in any detail. In general, they are as variable in the cranial 
index as those of the Middle and Late Aurignacian, although in western 
Europe the head form in this later period seems to run somewhat longer. 

Some of the skulls, as typified by the famous Ghancelade and by the 
male from Obercassel, show, however, something new a so-called Eski- 
moid modification of the masticatory apparatus. This consists of an 
even greater widening of the zygomatic arches than had been previously 
known; a flattening of the parietals, an enlargement of the area of tem- 
poral muscle attachment, and a keeling of the cranial vault. These fea- 
tures are accompanied, most markedly in the Chancelade specimen, by 

60 It is the modern tendency to deny African influences in the Spanish Upper Palaeo- 
lithic. Vaufrey (Anth, vol. 43, 1933, pp. 457-483) shows that the Capsian did not enter 
Spain, nor did it extend westward of Central Algeria. The Oranian was formerly called 
Ibero-Marusian until it was determined that this, too, was absent from Spain. Never- 
theless Spain was influenced, during the Upper, Palaeolithic, by some microlithic indus- 
try, which must have come from points south and east, of the same general type as that 
which went to Kenya as Wilton, to Egypt as Sebilian, to Palestine as Natufian, and to 
North Africa as Capsian and Oranian. 

81 Two Russian skulls from Undori may be Magdalenian and not Aurignacian. 
Talko-Hryncewicz, PAn, vol. 1, 1926, p. 208. 

Field, H., AA, vol. 38, 1936, p. 277. 

Pavlow, A., AnthPr, vol. 3, 1925. 

62 The Spanish material is particularly unsatisfactory. Dr. Obermaier in 1924 re- 
jected all previously studied finds except for two cranial vaults which had been cut down 
to serve as drinking bowls, a femur, and a few teeth. (Obermaier, H., Fossil Man in 
Spain, pp. 288290.) Two large series from Segovia, published by Dr. de las Barras de 
Aragon (AMSE, vol 12, 1933, pp. 90-123) as Magdalenian or Mesolithic, include one 
trephined skull. Our earliest positive case of trephination in Europe dates from the Late 
Neolithic. Furthermore, one of the sites contained pottery. 


a great eversion of the genial angles, a prominence of the malars, and a 
consequent flattening of part of the facial plane. 

This new adaptation, common among living Eskimos and Siberians, 
has been interpreted by a number of authors 63 to mean that the Mag- 
dalenians as exemplified by Chancelade were the ancestors of the living 
Eskimo, whose forebears moved northeastward as the ice retreated, and 
eventually crossed Behring Straits. But several objections have been 
raised to this identification. The nasal bones of Chancelade, in the first 
place, which were broken off and lost soon after the skull had been dis- 
covered, were very highly arched, projecting, and even hawk-like. 64 
They were thus extremely European in form, and not typically Eskimoid 
or mongoloid in the modern sense. 

On cultural grounds, Birket-Smith and Matthiassen have postulated 
that the Eskimos are not the product of a simple eastward migration from 
Asia, but that their origin is linked with that of the American Indian. 65 
It may be true that the similarities between Eskimo and Magdalenian 
culture are due to convergence, although this thesis has by no means 
been finally established, On physical grounds as well, evidence has been 
adduced to show that the Eskimo is really close to the American Indian. 66 

The question of Magdalenian-Eskimo relationships, in any case, is part 
of the general problem of those between Upper Palaeolithic man of the 
entire northern zone, and the origin of the American aborigines as a whole. 
It is too early at present to settle either. 

Returning to Ghancelade, we see that this individual differed in many 
ways from the standard Upper Palaeolithic mean. His face was very long, 
like that of Predmost; his orbits were high, and, like those of Combe 
Capelle, narrow. None of the other Magdalenian skulls which simulate 
him in " Eskimoid' 5 character diverge so completely from the total group, 
and hence Chancelade had been set apart by many as a separate race. 
Unlike the Late Aurignacians, he was a short man, about 160 cm. high, 
and his stature would be usual among most of the present inhabitants of 
the Arctic circle. His extremities, with his short heel bones, would also 
not be alien to the latter. 67 

Besides Chancelade and other individuals which approximate his type 
in varying degrees, the true Cr6-Magnon of Aurignacian tradition sur- 
vived unchanged into the Magdalenian. The Laugerie Basse cranium 

Morant, G. M., AE, vol. 1, 1926, pp. 257-276, is the latest and most exhaustive ex- 
position of this view. 

w We must thank Sir Arthur Keith for this discovery. An old and rare photograph, 
reproduced on page 395 of New Discoveries, establishes tfiis point definitely. 

Birket-Smith, K., PICA, 1930, pp. 470-475. 

M Shapiro, H. L., PSC, 1934, pp. 2723-2732; APAM, vol. 31, 1931, pp. 345-384. 

Seltzer, O. C,, HB, vol. 5, 1935, pp. 313-370. 

87 Bonin, Gt von, op. cit. 


could well fit into such a series. Others, such as the Le Placard cranium 
and skull C from Aveline's Hole in England, 68 represent an unreduced sur- 
vival of the brachycephalic eleiftent in the Cr6-Magnon complex. Still 
other skulls are smaller than the Upper Palaeolithic standard, show a 
reduction in browridges and in malars, and anticipate the general reduc- 
tion in size and in ruggedness which was to alter profoundly some branches 
of the Upper Palaeolithic stock in Europe and Asia after the close of the 
Pleistocene. It is important to learn that this reduction had already begun 
as early as the Magdalenian, and that at that time there was no geograph- 
ical difference between those which were and were not affected by this 
incipient tendency. 

During the Magdalenian, then, the internal diversity of Upper Pa- 
laeolithic European man became more noticeable than before. Some of 
the examples which are left to us represent a continuation of pre-existing 
Aurignacian forms, others show a modification found among living peoples 
of the Arctic, while still others anticipate the size reduction of the Mes- 
olithic. We may, if we like, attribute these differences to local segrega- 
tions and modifications, but since our knowledge of race in Magdalenian 
Europe covers so small a portion of the area in which that culture existed, 
it is perhaps more reasonable to postulate new movements as well as local 
survivals and changes. 


Our knowledge of the eastern distribution of Upper Palaeolithic man 
during the height of glaciation has been enormously extended by recent 
discoveries made by Dr. Pei, by the late Dr. Davidson Black, by P&re 
Dr. Teilhard de Ghardin, and by Dr. Franz Weidenreich, who have es- 
tablished the presence of several varieties of Late Pleistocene sapiens man, 
including the European type, in China and Mongolia. 69 

At Chou Kou Tien, close to Peiping, the discoverers of Sinanthropus have 
also found three well-preserved skulls, with one mandible and most of the 
accompanying long bones, in limestone pockets of late glacial debris, 
which includes Upper Palaeolithic implements analogous to European 
types. The preliminary descriptions of the cultural remains would suggest 
late rather than early Upper Pleistocene age. One of these skulls, the 
one with the mandible, seems, upon preliminary examination, to resemble 

68 The skulls from Aveline's Hole, Kent's Cavern, and Cough's Cave, were described 
by Keith, who considered them to be of Mesolithic age (Antiquity of Man, p. 407; New 
Discoveries, pp. 406-421); but Clark, an outstanding authority on the Mesolithic in 
northwestern Europe, indentifies them as Magdalenian (Clark, J. G. D., The Mesolithic 
Age in Britain, p. 107). 

69 The first discovery of this nature was of a sapiens tooth from the Sjara-osso-gol 
Deposits in Mongolia. Black, D., BGSC, vol. 5, 1927, p. 285. 

Also, Keith, Sir A., New Discoveries, pp. 250-251. 


the European Upper Palaeolithic group very closely, and especially the 
male of Qbercassel; it has also been compared to Ainu crania. A second 
skull greatly resembles that of a modern Eskimo, while a third may be 
compared to the racial type which invaded Japan during Neolithic times. 70 
The importance of these skulls cannot be overemphasized. They indi- 
cate that in eastern Asia as well as in Europe, the Late Palaeolithic group 
was already racially complex; that peoples of European type stretched 
across the entire width of the northern half of the Eurasiatic continent; 
and that the mongoloid family of races had already begun its character- 
istic development. By means of this knowledge we may explain, at least 
in part, the enigma of the Ainu, a large-headed, broad-faced white group 
living on the outer periphery of eastern Asia. At the same time fresh light 
is thrown upon the human materials which may have taken part in the 
early peopling of America. 


Although the Pleistocene men are long dead, and factory workers scurry 
to their labors where the Magdalenian hunters once impounded reindeer, 
the problems of human racial origins, and of human development during 
the PleistQcene, are still of great importance. On the foundations of our 
knowledge of Pleistocene man, in Europe, in Asia, and in Africa, must be 
built the interpretation of later and more complicated racial movements, 
racial survivals, genetic continuities and genetic changes. For this reason 
it seems better advised to state without trepidation the reconstruction of 
Pleistocene racial events which the facts themselves suggest, than to defer 
to more cautious and perhaps wiser opinions. 

These conclusions, which are by no means novel, 71 maybe stated briefly: 

(1) Homo sapiens was fully evolved as early as the mid-Pleistocene, if 
not earlier. 

(2) The earliest Homo sapiens known, as represented by several examples 
from Europe and Africa, was an ancestral long-headed white man of 
short stature and moderately great brain size. 

(3) The negro group probably evolved parallel to this white strain, 

to I am indebted for this information to Dr. Franz Weidenreich, who has given me 
permission to publish this preliminary notice. These comparisons are tentative and the 
reader must await Dr. Weidenreich's definitive publication for more detailed and more 
exact information. 

71 Aichel, Marett, and most recently Krogman, take stands essentially similar to the 

Aichel, O. s Der deutsche Mensch, pp. 12-36. 

Marett, J, R. de la H., Race, Sex, arid Environment, 

Krogman, W. M., "Cranial Types from Alishar Hiiyttk," in H. H. von der Osten, 
The Alishar Hiiyiik, Oriental Institute Publication #30, part IV, pp. 213-293. 


from a related sapiens ancestor. At what point the ancestors of negroes and 
whites diverged is not known. 

(4) During the Middle Pleistocene, if not at other times as well, a mix- 
ture took place between early white dolichocephals and one or more non- 
sapiens hominid species, including Homo neanderthalensis. 

(5) The result of this mixture was the development of a reasonably 
stable hybrid race, which was characterized by an excess of size, both of 
brain case and of bodily bulk. Although differing metrically from the 
rest of Homo sapiens as a whole, its character was nevertheless mainly 
sapiens, and only to a small extent Neanderthaloid or non-sapiens. Within 
the sapiens species, its relationship was with the whites. 

(6) This predominantly sapiens character may have been partly the 
result of convergent evolutionary tendencies on the part of the non-sapiens 

(7) Modern white men must include both individuals and racial en- 
tities which respectively possess and lack this non-sapiens strain, since all 
branches of the white stock did not mix with it. 

(8) On the basis of Palaeolithic cultural phenomena, one cannot assume 
that the non-sapiens element absorbed through mixture was less intelligent, 
or, in the social and intellectual sense, less human, than the original 
sapiens species. Modern European races which possess the former element 
show no signs of intellectual inferiority, or of any other discernible mental 

(9) Most if not all of the basic variations of bodily and cranial form, 
including brachycephaly, which occur among white men, already existed 
during the Late Pleistocene. The materials for the differentiation of 
white races and sub-races in post-glacial times were all present. 







Redrawn to scale: Fig. 4, from Keith, Sir A., The Antiquity of Man, Fig. 63, p. 188. 
Fig. 5, from Aichel, O., Der deutsche Mensch, Plate 2. Fig. 6, from Boule, M., and 
Vallois, H., AIPH, Mem. 18, 1937, Plate 14. 






Redrawn to scale: Fig. 7, from cast by J. H. McGregor, 1919. Fig. 8, from Keith, 
Sir A., and McCown, T. W., BASF, Bull. 13, 1937, Plates 5 and 6. Fig. 9, 
from Aichel, O., Der deutsche Mensch, Plate 4. 






Redrawn to scale: Fig. 10, from Boule, M., Vallois, H., and Vemeau, R., AIPH, 
Mem. 13, 1934, Plate 13. Fig. 11, from Aichel, O., Der deutsche Mewch, Plate 17; also 
Kossinna, G,, Urspntng und Verbreitung der Indogermanen, Fig. 134, p. 123. Fig. 12, 
from Retzius G., Crania Suecica Antiqua, Plates 39-40. 






Redrawn to scale: Fig. 13, from Vallois, H., Anth, vol. 40, 1930, Fig. 2, p. 344. 
Fig. 14, from Crania Britannica, vol. 2, Plate 59. Fig. 15, from Kossinna, G., Ursprung 
und Verbreltung der Indogermanen, Fig. 102, p. 90. 

Chapter III 


The Mesolithic cultural period, which follows the final Palaeolithic in 
Europe, is wholly post-Pleistocene in that continent, and extends roughly 
from immediately post-glacial time to 3000 B.C. and later. 

The Mesolithic manner of living was primarily similar to that of the 
Upper Palaeolithic. People still relied on hunting and the gathering of 
wild vegetable products for food, and the population must have remained 
as sparse as ever. Man had acquired but one domestic animal the dog, 
which may have helped in hunting, but which was not bred for eating, 
and hence served as only an indirect source of food. The Mesolithic econ- 
omy was, therefore, a prolongation of the Upper Palaeolithic system into 
relatively recent times; in the technical sense, however, there were certain 
improvements; with the introduction of microliths composite weapons 
were made; dugout canoes furnished good water transportation, and tree- 
felling axes must have made the building of adequate houses possible. 
The forerunners of the textile arts were probably developed to permit the 
manufacture and use of fish nets, good basketry, and matting. 

The cultures of the Mesolithic period in Europe may be divided into 
two elements of different origins, which in many regions met and blended. 
One was the intrusive Tardenoisian with its advanced microlithic tech- 
nique, which came in from the south across the straits of Gibraltar, and 
perhaps around the eastern end of the Mediterranean. 1 These migrations 
into Europe from the south were caused by climatic shifts incident upon 
the final glacial retreat. As the glacier moved northward to take up its 
last stand in the high Scandinavian land-mass, the erstwhile well-watered 
and temperate belts of North Africa and the Near East suffered a gradual 
desiccation. As the rain-belt moved northward, zones of temperate and 
sub-tropical climate shifted from Africa to southern and central Europe, 
and the climate of Europe became warmer in early post-glacial times than 
it is at present. The people who brought the elements of the Tardenoisian 
complex northward had been accustomed to hunting on open grasslands 
before their arrival in Europe, and they, therefore, settled in sandy regions 

1 Clarke mentions this second route as a possibility. Clarke, J. G. D., The Mesolithic 
Settlement of Northern Europe, pp. xiv xv. 



and treeless highlands, since neither their tool kit nor their general manner 
of living was suited to a forest environment. 

The second cultural element was furnished by the survival of the old 
Upper Palaeolithic techniques, employed by the descendants of the rein- 
deer hunters. The gradual growth of forest in what had formerly been the 
North European tundra belt forced them to learn a new kind of hunting 
and to live on the flesh of new animals, while the warming of northern 
waters gave them an abundance of fish and molluscs, focussing their 
attention not only on the forest but also on the rivers and sea. 

In the north and west of Europe, where the glacier lasted the longest, 
cultures of Aurignacian and Magdalenian tradition survived into the full 
Mesolithic, when some of them blended in varying degrees with the newly 
arrived Tardenoisian. In outlying regions, such as the north coast of 
Ireland and Finnmark in Norway, flint implements of Upper Palaeolithic 
inspiration may still have been made as late as the time of Christ. 

For the purpose of simplification, therefore, the history of the Mesolithic 
period in Europe may be reduced to two elements: (1) an invasion of 
microlith-makers from southern regions which had been temperate and de- 
sirable during the Late Pleistocene, but which were now drying up and be- 
coming less habitable than Europe; (2) survival of the Palaeolithic people 
of Europe in various regions and in varying intensity, but concentrated es- 
pecially in the northern forest belt, along the western coasts, and in the 
centers where the ice had lasted longest, notably, Norway and Switzerland. 


Before gathering information which will help us in Europe, let us first 
see what changes or continuities occur in Africa, with the passage into 
post-Pleistocene time. In East Africa Leakey has found skeletons asso- 
ciated with a microlithic culture which he calls Elmenteitan, probably at 
least partly contemporary with the post-glacial Mesolithic cultures of 
Europe, 2 with which he has tentatively correlated it. The series includes 
the skulls of three adult males, three adult females, and one child, as 
well as a number of miscellaneous long bones. The bodies which they 
represent had been placed in rock niches on either side of a watercourse, 
and a subsequent flood had washed most of them out and deposited them 
in silt. Hence, it is impossible to associate the long bones with the crania. 

From this series of seven skulls, it is evident that the earlier East African 
Mediterranean racial types were carried over into the post-glacial period 
with little or no alteration. The vaults of the Mesolithic skulls are again 
comparable in size to Galley Hill and Combe Capelle. The shape of 
these vaults, however, is now variable, at least in the female group, for 

2 Leakey, L. S. B., Stone Age Races of Kenya, Chapter 6. 


one of the latter skulls has a cranial index of 80. This sex distinction in 
shape, may, of course, have been equally present in the Aurignacian 
group, but we have no material to confirm it. 

The face continues the special evolution which had begun during the 
Aurignacian; it grows both longer and broader, while the nasal height 
increases at the same time. 3 Both the faces and noses of these skulls are 
exceptionally long, by any racial standards. All of them have high orbits. 
The nose remains leptorrhine, but the nasal skeleton is not highly arched; 
some of the crania, especially one of the female specimens, show consider- 
able prognathism. In general, the foreheads are sloping, the browridges 
and other bony markings slight to medium on the males, while on the 
females the browridges are actually lacking. One male specimen, 
Elmenteita A, differs from the others; the mandible has everted gonial 
angles while on the cranium the temporal crests rise high over the parietals, 
producing a narrow forehead, and giving the whole head a pseudo- 
Eskimoid appearance. 

In the Gamble's Cave Aurignacian series, since only males were repre- 
sented, it was impossible to tell whether or not any extensive differences 
between the sexes existed. In the Elmenteitan group, the male crania ex- 
ceed the female ones considerably in the length and breadth of the vault 
and in the face heights and face breadth. Vault heights, forehead breadths, 
and orbital dimensions are much the same in both, however. These East 
Africans, therefore, while lacking the bony luxuriance of the Upper Pa- 
laeolithic Europeans and North Africans, do exhibit a positive amount of 
sex linkage in the characters which make them racially distinctive. 

Despite the continued residence of this long-faced racial group in East 
Africa, there is still little that is negroid about most of the skulls. The 
forehead, in some of the females, is a bit bulbous, but so it is with many 
living Mediterraneans; some of the jaws project forward with a consid- 
erable alveolar prognathism, but so do the jaws of a number of early 
European crania. 4 The nasal index, which falls near the human minimum, 
is at the opposite extreme from those of negroes. The nasal bones, present 
in but two crania, are long, narrow, and hour-glass shaped; they taper 
upwards, and penetrate high into the frontal bone, as with certain anthro- 
poid apes and the Eskimo; but the two bones are not greatly arched and 
the nasal vault, in these two specimens, is low. Thus the nasal bones pos- 
sess an individual character which is neither typically white nor negroid. 

The Elmenteitan people remained as tall as the Upper Aurignacians. 

3 Nasion-menton heights on males are 126 and 132 mm.; nasion-alveon 80 and 
81 mm.; nose heights 58 to 59 mm. 

4 The female skull #F 1 is the most nearly negroid of all, and in this case a definite 
negro strain seems very likely. 


The mean stature for six males is 178.7 cm., for three females, 152.5 cm.; 
thus, the sex differences are great in bodily size, as well as in head and face 
diameters. The greater stature and sex differentiation of these East Afri- 
cans may have been simply the result of evolutionary change; one cannot 
find a non-sapiens species to provide these modifications, as in the case of 
the Upper Palaeolithic Europeans. 

Before we leave East Africa for some time, it may be interesting for us 
to note that Leakey has also found a number of skeletons associated with 
a parallel culture, Wilton A., located nearby and probably not later than 
Elmenteita. These Wiltonians were tall, heavy-boned men, with large, 
strongly arched foreheads, and small faces, very much like the Strand- 
loopers from South African shell heaps, and ancestral Bushmen. Thus 
along the Lake Victoria shore line, not far away from Elmenteita, were 
ancestral Bushmen, living in geographical proximity to Mesolithic ances- 
tral Hamites. The East African whites lived on a racial frontier, and not 
in a center of white racial differentiation. If Bushmen traits turn up now 
and then among Hamites or Hamitic traits among Hottentots or Bushmen, 
there is little wonder. 

North Africa was occupied, during the post-glacial Mesolithic period, 
by the Middle Capsian successors of the Afalou people. These are known 
through a collection of skulls from the site of Mechta el 'Arbi, of which 
only nine have been studied in any detail. 6 All come from what Aram- 
bourg calls the Middle Capsian, which has been correlated chronologically 
with the European Solutrean by Menghin, with the Solutreo-Magdalenian 
by Obermaier, and with the Mesolithic by Vaufrey. 6 They are considered 
here rather than in the preceding chapter since they belong with the 
Mesolithic in the European sense both racially and culturally, whatever 
their chronological position. 

It is impossible, unfortunately, to treat these skulls with complete clarity. 
Judging by published measurements, photographs, and drawings, we 
may conclude that on the whole they resemble the earlier Afalou skulls in 
a general way, but that most of them are smaller and lack the ruggedness 
of their predecessors, having weaker browridges, less pronounced muscular 
markings, and narrower faces. Some of them have vertical foreheads, a 
feature foreign to the Afalou people. They still retain in most instances, 
however, a low face and low orbits, and a range of head form reaching 
the limits of the earlier series. 

8 Probably over fifty crania have been removed from this site by successive expedi- 
tions, but only five have been carefully studied. See Cole, Fay Cooper, LMB, vol. 1, 
1928, pp. 167-189. Four others, of which two only are from the Mechta site, have been 
dealt with, as thoroughly as the data permitted, in the Afalou volume. 

6 Vaufrey denies the existence of a Middle Capsian, and says that these skulls are Late 
Capsian, which he considers Mesolithic, 


In their degree of size reduction, and diminution of sex-linked bony 
profusion, they may be likened to some of the Mesolithic crania from 
Europe, which will be studied later in this chapter. It is quite likely, as 
Cole suggests, that one of the Mechta skulls showed a negroid tendency, 
while the others were subjected to mixture with Mediterranean racial 
elements. The inference is that the countries at the eastern end of the 
Mediterranean, from which these influences probably seeped, were already 
inhabited by small Mediterraneans. On archaeological grounds, it is 
unlikely that these Mediterranean racial elements came directly from the 

Our entire knowledge of the racial composition of the early inhabitants 
of the southern Sahara is furnished by a single skeleton, unearthed at 
Asselar, a military post some four hundred kilometers north of Timbuctu, 
in what is now utter desert, 7 but what was at the time a fertile, well-watered 
plateau, drained by wide rivers, and rich in grass and ruminant game. 

The skeleton, which had not been buried but which simply lay in the 
place of death, was covered by lake-laid sands. These same sands have 
yielded the bones of huge fish, in the same state of fossilization as those 
of the man, and the shells of fresh-water molluscs, which indicate that the 
region of Asselar was at that time still a lake country, with running streams 
and a forest border, near the southern limit of the south-Saharan grass- 
lands, and the northernmost extension of the tropical forest. Asselar man 
died before this region had become desiccated, but his cultural association 
is Mesolithic or Early Neolithic, and his chronological age unquestionably 

He was a tall man, over 170 cm. in height; his limbs were long in pro- 
portion to his trunk, and his forearms and lower legs long when com- 
pared to the proximal segments of their extremities. His hands were long 
and slim, with small carpal bones, unlike the broad hands and thick 
wrists of the Afalou men farther north. 

The skull is of medium vault size, comparable to Grimaldi, Afalou #28, 
and the Kenya Aurignacians. Like all of these, it is dolichocephalic, 
with a cranial index of 71 . The muscular markings of the vault are slight, 
and the browridges weak. In facial dimensions, Asselar is intermediate 
between the Grimaldi and East African extremes. Morphologically, how- 
ever, it is the most negroid specimen of equal age yet found. The malars 
project forwards, and the lower border of the orbit stands in front of the 
upper, when the skull is placed in the eye-ear plane. The nose is chamaer- 
rhine, and negroid in conformation. 

Asselar man was either an incompletely evolved negroid, or a negro- 

*Boule, M., and Vallois, H. V., AEPH, Mem. 9, 1932. 
See also Bailly, Rene, RA, vol. 43, 1933, pp. 172-181. 


white hybrid; he did not closely anticipate, in cranial form, the modern 
blacks from the Guinea Coast and Sudan. He retained certain tendencies 
in a white direction, and others which related him to the Bushmen and 
Hottentots. The Asselar find, like those in East Africa, makes it very likely 
that the spread of fully differentiated negroes into much of their present 
area in Africa was a fairly recent phenomenon. 


Compared with the continent of Africa, from the prehistoric standpoint, 
Asia is archaeologically little known. So far, excavations have revealed 
implements of Mesolithic technique in Kurdistan and in Palestine, 8 but 
only from the latter have Mesolithic skeletons been recovered. Here an 
Aurignacian culture lasted during the entire Late Pleistocene, and directly 
preceded the Mesolithic. Since Miss Garrod feels that this region was one 
of the main areas of differentiation of the Aurignacian cultural technique, 
it is very unfortunate that not a single Aurignacian skull has been pub- 
lished. Therefore, the very important question of the Late Pleistocene re- 
lationships of this key area must remain unsettled. 

For the following period, however, at least two hundred skeletons have 
been exhumed from two different Mesolithic levels and from five or more 
sites. So far, only two of these skeletons have been published, one from 
each level. Great doubt is current at the moment concerning the exact 
nature of the physical types of this people, and we must await detailed 
publications in the near future before this matter may be settled. 9 

These Palestinians, who have been given the name Natufians, appar- 
ently differed in physical type from period to period. One of the two skele- 
tons which has been published is that of an adult female from the earliest 
level at a site called Erg el Ahmar. 10 

The skull of this woman is large, robust, and thick- walled; it is purely 
dolichocephalic, and has an elevated cranial vault in which the height 
almost equals the breadth. The forehead, as with females of many races, 
is broad, straight, and rounded. The* face, likewise, is broad, and of 
medium height; the nasal root, somewhat depressed, is hidden under 
browridges massive for a female, while the nasal bones project far forward, 
to form an accentuated profile. 

The low, broad orbits of this specimen assume the rectangular form 
characteristic among most of the Upper Palaeolithic skulls from Europe 
and North Africa, while the orbital index is correspondingly low. The 

8 Garrod, Miss D. A. E., BASF, No. 6, 1930, pp. 8-43. 

9 Mr. T. D. McCown was, at the time of writing, engaged in working over a large col- 
lection of these skeletons under the direction of Sir Arthur Keith and intends to publish 
it shortly. 

10 Vallois, Henri, V., Anth, vol. 46, 1936, pp. 529-543. 


nose is high, narrow, and metrically leptorrhine; the nasal spine promi- 
nent, and the lower border of the piriform opening strongly crested. The 
mandible, of medium robust icity, possesses a prominent chin. The rugged 
beauty of this Natufian woman was, however, somewhat diminished by 
an abnormality of dental occlusion, for her lower incisors overlap the 
upper ones. 

Morphologically, this skull is perfectly European and belongs without 
question to the general Upper Palaeolithic type. It would also fit metri- 
cally into the female range for this group. It would, however, fit equally 
well into the North African series of Afalou bou Rummel, except that it is 
somewhat narrower nosed than the females of that group as known at 
present. 11 In the absence of data on Palestinian Aurignacian crania, one 
may suppose that the Aurignacian Upper Palaeolithic Neanderthal- 
sapiens hybrid developed in this neighborhood from Skhul-like beginnings, 
and that this Erg el Ahmar female is a survival of it. 

The skulls from the later Natufian period, while exceedingly numerous, 
remain dubiously classified because of several conflicting ideas about them 
which have been published. Sir Arthur Keith 12 in a preliminary report on 
the remains from Shuqbah and Kebara, states that the later Natufians 
were short people, the males having a mean stature of 160 cm. and the 
females of 152 cm. The tallest male in the group was only 165 cm. in 
height. The hands and feet of these later Natufians were remarkably 
small, and their long bones were in no sense massive. 

The skulls which Keith describes are of a peculiarly Mediterranean 
type, with a cephalic index ranging from 72 to 78, thus rivalling the sub- 
dolichocephalic head form of short statured Mediterraneans living today. 
The brain cases are of medium size, and the faces absolutely small. The 
lower jaws are also small and weakly developed, with little chin promi- 
nence and a prevalence of alveolar prognathism. The wide, low- vaulted 
nose, in combination with prognathism, gives a somewhat negroid cast 
to the face. The browridges are smooth, and the whole system of muscular- 
ity in the male but slightly developed. These late Natufians represent a 
basically Mediterranean type with minor negroid affinities. 13 There was, 
apparently, a change of race during the Natufian. These small Mediter- 
raneans must have brought their microliths from some point farther south 
or east, impelled by changes of climate. 

11 Some of the Mugharet el Wad crania, which belong to the earlier horizon, seem 
likewise to resemble those of the Upper Pleistocene. This comparison represents, how- 
ever, a preliminary impression, and is stated only with reservations. Personal com- 
munication by Mr. T. D. McCown. 

** Keith, Sir A., New Discoveries, pp. 202-214; PICP, 1932, pp. 46-47. 

13 This impression is also confirmed by the French school. 

Boule, Vallois, and Verneau, Les Grottes Palaeolithiques de Bent Seghoual, pp. 212-214. 



Although, during the last century, many skulls have been removed from 
caves in various parts of Spain, not one of them may be assigned with 
complete security to the Mesolithic period. Since Spain was apparently 
the main if not the only highroad of migration northward from Africa 
into Europe during the Mesolithic, this gap in our knowledge is ex- 
tremely unfortunate, particularly in view of the parallel deficiency in 

Late Mesolithic skeletons have, however, been found in Portugal, in a 
series of shell-heaps which lie on a raised shore near the village of Muge, on 
the eastern bank of the Tagus River, some fifteen miles upstream from the 
head of tide-water. At the time of occupation, the shellfish which the 
midden-builders ate lived in salt water, 14 and the land must have lain 
several meters lower than its present level. This sinking may probably be 
correlated with the formation of the Litorina Sea, which lasted in what is 
now the Baltic from 5600 to 2500 B.C. If this dating applies to Portugal, 
the Muge middens were probably formed nearer the end of this period 
than its beginning. The safest dating for this site is immediately pre- 
Neolithic, 15 if not early Neolithic, in the third millennium B.C. 

Over two hundred human skeletons have been removed from these 
middens at various times during the last eighty years. Of this number, 
however, only nine have been measured and published in such a way that 
we may profitably consider them here. 16 In the past, many curious ideas 
have been circulated about the racial types represented by these remains, 
and these notions have been widely credited and frequently repeated. The 
principal misconception has been that the Muge crania include two types: 
a non-European negroid, and a hyperbrachycephal variously called Alpine 
and mongoloid. 

Actually, there is no evidence to show among them a greater negroid 
tendency than is commonly found among many living Europeans of 
Mediterranean extraction, while the so-called "brachycephalic" skulls are 
probably all or almost all mesocephalic, since some were badly warped 

14 Obermaier, H., Fossil Man in Spain, p. 325. 

16 Obermaier, op. cit., p. 325, says: "The fauna of these deposits does not include 
any domestic animals except perhaps the dog and consists of wild cattle, deer, sheep 
or goat, horse, swine, dog or wolf, felines, badger, civet, and hare." (Italics are mine.) 
The Iberian Peninsula is not known, at the period in question, to have sheltered either 
wild sheep or wild goats. The only animal which could possibly have been mistaken 
for either is a diminutive ibex, the bones of which are much smaller than those of either 
sheep or goat. Unless the bones in question are actually those of ibex, the Muge midden- 
dwellers must have already met the first waves of the Neolithic economy from North 
Africa. Agriculture and domestic animals did not necessarily enter the Iberian Penin- 
sula in one magnificent sweep; scattered families of herdsmen may have wandered over 
as an advance guard. 

16 Vallois, Henri V., Anth, vol. 40, 1930, pp. 337-389. 


by earth pressure, and others were improperly measured; while still others 

have been lost or mislaid. l7 

The cranial series from Muge, as it is known at present, is reasonably 
homogeneous. The cranial index ranges from 69 to 80, or possibly 82, with 
most of the skulls in the low seventies. One may postulate a mean of 
about 75 to 77 on the living, The brain case is of medium size, but rela- 
tively high; ovoid in form, flattish on the top, and gently rounded in the 
occipital region. The female crania have vertical foreheads, while those 
of the males are sloping; the frontal bone in both is always strongly curved. 
On most of the male skulls, the browridges are well developed in the 
median segment, but not on the sides, while on the female specimens, the 
supraorbital region is usually quite smooth. 

The orbits are low, but not especially narrow. The nasal dimensions are 
small, yielding a mesorrhine index; the lower border of the nasal opening 
is usually sharp, but in some cases it is rounded, and in one guttered. The 
face is mesoprosopic, being both low and extremely narrow. In the photo- 
graphs, the zygomatic arches appear to be delicate, and closely aligned to 
the temporals. The mandibles are of moderate height, but narrow, while 
the palates are quite large for the total size of the skulls, and the teeth are 
also large. Most of the skulls show a slight alveolar prognathism, which in 
a few instances is quite marked. 

Among the nine crania measured by Vallois, the females equal the 
males, or approach them closely, in all dimensions. Sex differentiation, 
therefore, is practically absent from the metrical standpoint, but the differ- 
ence in browridge development is apparently sufficient to permit the crani- 
ologist to distinguish them readily. The long bones, studied apart from 
the crania, to which they cannot be matched, give a reconstructed stat- 
ure of 160 cm. for the males, and 152 cm. for females. Despite this short 
stature, a limb form found among some Upper Palaeolithic peoples is re- 
peated here the distal segments are long when compared to the proximal. 

The racial position of the Muge population cannot be finally deter- 
mined until more evidence, both internal and comparative, is at hand. 
Yet from present indications there seems every reason to believe that the 
Portuguese midden-dwellers were very similar to, or identical with, the 
late Natufians of Palestine, and that both represented a northward thrust 
from a Mediterranean racial homeland somewhere in southwestern Asia, 
northeastern Africa, or both. 


Our knowledge of Mesolithic man in France is little better than that 
of the Iberian Peninsula, despite the extensive digging which has been 
17 Vallois, op. cit. 


going on there for almost a century, French Mesolithic sites are di- 
vided into two main cultural groups, the Azilian and the Tardenoisian. 
The Tardenoisian represents the northward advance of the Capsians 
from North Africa, and its eastward spread across central Europe to 
Russia, and perhaps beyond. The Azilian represents a degenerate Mag- 
dalenian cultural expression surviving in southwestern France, in the 
Asturias of Spain, and in parts of England, under incoming Tarde- 
noisian influence. By the time of the full Mesolithic, the fauna of 
France had changed almost completely, for the reindeer which the 
Magdalenian people had hunted had been replaced by red deer, while 
the impressive mammoths and other large mammals were by now long 

The only French Mesolithic series known, aside from single skeletons, 
comes from T6viec, a small island to the west of the peninsula of Quiberon, 
Morbihan, Brittany. 18 Here a coastal population subsisted on molluscs, in- 
cluding Litorina, and crustaceans, with little hunting. Its remains, consist- 
ing of twenty-one skeletons, come from stone cists buried in a midden on a 
raised beach. The implements, as shown in the archaeological part of the 
T6viec report, seem to be of a marginal, Azilian-like Epipalaeolithic char- 
acter, like those from the Asturian horizon in Spain, with some late Tar- 
denoisian influence. On the basis of the artefacts, the raised beach, the 
Litorina skulls, and the stone cists, one must suppose that the remains 
cannot be older than the fourth millennium B.C., and may be even later. 
However, they are purely Mesolithic and antedate the local Neolithic, 
however retarded. 

Of the twenty-one skeletons, seven adult males and eight adult females 
have been studied (see Appendix I, col. 3). The skulls are reasonably uni- 
form; they are smaller in size than the Upper Palaeolithic French crania, 
but a little larger than those of the Muge people or Natufians; the vault is 
as high as its breadth; the cranial form between dolicho- and mesocephaly, 
with a male mean of 74.3, and the narrow range of 72 to 77. These skulls 
are thick boned, and rather massive in structure. Morphologically, they 
resemble the Upper Palaeolithic rather than the Mediterranean form. The 
faces are low and relatively broad, with the bizygomatic diameter often 
exceeding the head breadth. The browridges of some of the males are 
rather heavy, the nasion depression deep. The noses are mesorrhine, and 
fully European in form; the orbits are low. 

On the whole, these skulls look like smaller replicas of Aurignacian and 
Magdalenian forms, or an intermediate stage between these and the 
Mediterraneans from farther south, as exemplified by the Portuguese and 

18 P6quart, Marthe, and St. Juste; also, Boule, M., and Vallois, H., AIPH, Mem. 18, 


Palestinian specimens. One skull in particular bears a striking resemblance 
to Chancelade. 

The statures of these people were low: 159 cm. for the men, 151 cm. for 
the women; the long bones not very heavy. The distal extremities were 
relatively long; not as much so as in some earlier skeletons, but more so 
than among most living Europeans. 

Our interpretation of these late Mesolithic remains from the western 
corner of France is that they represent a group of marginal Epipalaeolithic 
survivors from beyond the Pyrenees, pushed northward, partly by climatic 
changes and partly by the arrival of new people from North Africa. We 
have, after all, no other evidence to show us what kind of people inhabited 
the Iberian Peninsula during the Late Pleistocene; a conglomerate of first 
wave Grimaldi- Combe Gapelle-like Aurignacians plus some Magdalenians, 
plus some bringers of microliths from the south and east, would presum- 
ably look very much like this T6viec type, especially since the overgrown 
Middle and Late Aurignacians did not hunt south of the Pyrenees. 

Aside from this T6viec series, the Azilian culture proper is represented 
by the remains of four individuals removed from the Trou Violet at 
Montardit, 19 Arrifege, in the northern Pyrenees, near the type station of 
the Azilian at Mas d'Azil, and one mandible and several long bones, from 
Mas d'Azil itself. 20 Only one specimen, the so-called Montardit I, includes 
a complete skeleton, or even a complete brain case. Without elaboration, 
one may say that in every respect they belong to the same type as that of 


The third reasonably large series of Mesolithic crania in Europe comes 
from the Ofnet cave near Hohlheim in Bavaria, 21 where thirty-three skulls 
were found neatly arranged in a solid circle, like eggs in a nest. Nineteen 
of them belonged to children, ten to women, and only four to men. Along 
with the skulls were, in most cases, the two topmost neck vertebrae, the 
axis and atlas. The bodies were missing. A few miles away, at Kauferts- 
berg, a single adult male skull has been discovered which was buried under 
identical circumstances. 

Unfortunately, it is impossible to detect the murderers in this Mesolithic 
mystery. The peculiar sex ratio, the fact that all the heads were buried at 
once and while still fresh, and the further fact that all had been fractured 
by sharp blows with a lens-shaped implement similar in form to a round- 
poled celt, 22 make it unlikely that this was a normal, peaceful form of 

19 Sawtelie, R. O. (Mrs. Wallis), PMP, vol. 11, #4, 1931. 

20 Piette, E., BSAP, ser. 4, vol. 6, 1895, pp. 485-486. 

21 Scheldt, W., Die etszeitlichen Schadelfunde aus der grossen Ofnet-hohle y etc. 

22 Mollison, T., AAnz, vol. 13, 1936, pp. 79-88. * 


burial. The skulls were daubed with red ochre, and a few flint implements 
were left with them. From these clues we may deduce that the killers, 
their victims, or both, were culturally either Tardenoisians, 23 or Azilians, 
and that the date is probably Period II of the Mesolithic. 24 These skulls 
do not form a homogeneous group, but differ greatly in head form, as well 
as in other characters. Of the two dolichocephals, the male #K1818 is 
obviously an Upper Palaeolithic survivor, without visible change. This 
skull exceeds the rest of the series in general size. It is extremely massive, 
with projecting browridges, a retreating forehead, a very broad face, ex- 
tremely low orbits, and a ponderous mandible. Its cubic capacity is well 
over 1600 cc. Its companion, a male of lesser dimensions, is less extremely 
developed and falls closer to the general type of the series. 

The mesocephals, which include Kaufertsberg, are smaller, and rela- 
tively higher vaulted; as Boule and Vallois have pointed out, 25 they verge 
on the type of Teviec, which these authors consider to be Magdalenian 
survivals. At any rate, they do not seem to be intermediate between the 
dolichocephals just described, which resemble rather a full Aurignacian 
prototype, and the brachycephals. 

The last present the real Ofnet problem. Two of the brachycephalic 
crania are masculine, 26 two feminine, with the highest index, that of a 
female, 87. These skulls are long, wide, and of moderate vault height; the 
faces are without exception wide. In one of the male specimens (K1809) 
the greatest length of the skull lies in the forward segment, as with modern 
planoccipitals, such as Armenoids. The forehead of this skull is very wide, 
and the face extremely broad and low. The face is, furthermore, com- 
pletely orthognathous, and the lower jaw is very massive, with flaring 
gonial angles and a square, bilateral chin. This jaw is also very high, and 
reduces the impression of shortness in the total facial plane. The other 
male specimen, on the borderline of brachycephaly (K1800) is the only 
one in the whole group which is hypsiconch, and one of two that are lep- 

The female skulls show a considerable sex difference in head form, and 
likewise in browridges and other manifestations of bony relief. As with 
their Upper Palaeolithic prototypes, they are notably smaller, in most 
cases, than the males. On the whole, they vary much less than do the 
masculine crania, and fall closest to the brachycephalic male, K1809, in 

23 Clark, J. G. D., The Mesolithic Settlement of Northern Europe, p. 218. 

24 Mollison, in view of the cross-section of the implement with which they were killed, 
suggests that the date may have been Late Magdalenian, since no such implement re- 
appears until the Late Neolithic. Mollison, T,, op. cit. 

26 Boule, M., and Vallois, H., AIPH, Mem. 18, pp. 170-177. 
w Including #K1800, the C. I. of which is 79.85. 


Despite the differences between these skulls, which have been empha- 
sized in the foregoing description, they all have in common an Upper 
Palaeolithic character. The dolichocephalic males might have been direct 
descendants of the local Late Pleistocene population or intruders of similar 
type from North Africa. The mesocephals might have been, and probably 
were, the bearers of the Azilian culture from southern France to Bavaria. 
The brachycephals, on the other hand, may have been survivors of the 
type represented by Solutr6 #2 several thousand years earlier, but their 
resemblance to the brachycephals of Afalou bou Rummel is much stronger. 

It is possible that the old Afalou type was thrust into central Europe 
at the head of the wave of migration which brought the smaller Mediter- 
raneans to Portugal. It is likewise possible, but on archaeological grounds 
still impossible to demonstrate, that these brachycephals came from Asia 
Minor or Palestine, where an Afalou-like type existed in the early Natu- 
fian, and presumably still earlier. The question of the origin of these 
brachycephals cannot be settled without further data. 27 At any rate, the 
skulls of the Ofnet victims serve to show that various survivors of an older 
order had begun to assemble north of the Alps in early post-glacial times. 


From Ofnet eastward, Mesolithic Europe is a blank until we reach the 
Crimea, for in all the intervening territory, no human remains of this 
period have been found and described. In the small cave of Murzak 
Koba, 28 near the Crimean village of Chorgun, Soviet archaeologists have 
found two skeletons crushed under heavy stones which fell from the roof. 
One is masculine, the other feminine. 

The man from Murzak Koba was tall (180 cm.); long headed, large 
headed, with heavy browridges, and heavy superior oblique ridges on the 
occiput. The configuration of the nasion region, with a depressed root, 
is typical of Upper Palaeolithic European man; while the orbits are ex- 
tremely low and wide. The face is wider than the Upper Palaeolithic 
mean, and longer. Murzak Koba man was, without any question, a sur- 
vival of the eastern European Upper Palaeolithic type into Mesolithic 
times. His female companion apparently represented the feminine version 
of the same race. 

87 The photographs of at least three out of eight Ofnet crania, published by Scheldt, 
show apparent signs of tooth knocking. These are K1809, upper right lateral incisor; 
K1813, lower right median incisor; Kaufertsberg, lower left lateral incisor. Since in- 
cisor evulsion was found in some of the Natufian skulls, as well as in the one published 
Late Capsian cranium, this might indicate either a southeastern or a southwestern 
connection. (Personal communication of T. D. McCown, for Natufians. The Late Cap- 
sian skull is A!n Mlila, described by Boule, Vallois, and Verneau in the Afalou mono- 

* 8 Field, Henry, AA, vol. 39, 1937, p. 468. 


The cultural equipment of this couple consisted of flint implements and 
bone harpoons of Mesolithic type. The bearing of these two skeletons on 
the general problem of Mesolithic diffusion is that an entry of Mesolithic 
culture and race into Europe from a point this far east is rendered a little 
less likely. 

This does not, of course, affect in any way the possibility of an entrance 
by way of the Balkans; but extensive reconnaissance work in Jugoslavia 
has failed to locate any Mesolithic horizon there. 29 We are again forced to 
the conclusion that the bulk of the Mesolithic influences, both cultural and 
physical, which entered Europe after the retreat of the glacier came into 
that continent over the Straits of Gibraltar, 30 a conclusion which may, of 
course, need revision as new evidence shall come to light. 81 


During the maximum of the Wurm II glaciation, Palaeolithic man 
adapted himself, as we have seen, to the cold. Once so adapted it is 
only natural to suppose that he followed the climate to which he was ac- 
customed northward as the ice melted, and as the animals which he ate 
moved in the same direction. Now the final glaciation of Europe was cen- 
tered not in the most northerly point of the continent, but in the land 
mass of the Scandinavian Peninsula, and, apparently, the coasts of Norway 
(northern and western) were not affected by the last glacial advance. 
Human beings thus could penetrate Scandinavia in Late Palaeolithic 
times. 32 Scandinavia, therefore, formed the last retreat of Old Stone Age 
man fleeing before the encroaching temperate conditions with their growth 
of forest and change of fauna; for when the ice disappeared he had no 
further place to go. Here at last the changing climate of the early post- 
glacial period faced him with the final necessity of modifying his culture. 

The accompanying chart renders it unnecessary to describe at length 
the changes which took place in the Baltic region during the Mesolithic. 88 
In Period I, which lasted from the end of Palaeolithic times until about 
6800 B.C., following the de Geer chronolo'gy, a number of local industries 

a>Fewkes, V. J. } BASP, #9, 1933, pp. 17-32. 

30 Italy must be rejected as a likely avenue of entry, since the culture of that penin- 
sula during the entire Upper Palaeolithic was the Grimaldian, a local form of Aurigna- 
cian, which persisted, without a Solutrean or Magdalenian interlude, until Mesolithic 
times, and even through them to the Neolithic, with only a minor microlithic influence. 
Greenlee, R. F., The Association and Interrelation of the Microlithic Cultures of Europe and 
Africa (privately printed), 1935, pp. 28-31. 

31 Bonch-Osmolovskii, G., reports the discovery of another Mesolithic skeleton from 
the Crimea, buried in a crouching position. He states that it is earlier than Murzak 
Koba and is not of "Cr6-Magnon" type. Quoted by Field, H., AA, vol. 39, 1937, p. 467. 

82 Boe, J., and Nummedal, A., Le Finnmarkien. 

* Most of the following is based on Clarke, J. G. D., The Mesolithic Settlement oj North- 
ern Europe; also, Antiquity, vol. 12, 1938, pp. 154-171. 



developed in northern Europe from Palaeolithic origins. Thus the tanged- 
point cultures which stretched from Belgium to the Ukraine, across the 
north of Germany and of Poland, are derived ultimately from Aurignacian 





(after Blytt 
and Sernander 



(after Munthe 
and Saoramo) 


(after von Post] 



(after Clark) 












!l I 5 




^armer, moist, 
and oceanic) 



C Litortna 

Alder . 


3- </> 



1 | 


- 1 Maximum 
Bemmn4 of 




(warm, dry. and 




Hazel Scrub 
Birch and Pine 


^ * 





appearance of 
true forest) 

and Grasses 


POINT I Remouchamps, Swiderian, 
CULTURE JKomsa.Fosna 










Dryas Flora 


FIG. 16 

Adapted and combined from the charts of Godwin and Clark, with further modifica- 
tions suggested by Movius. Godwin, H., "Pollenanalysis," etc., The New Phytologist, 
Vol. 33, #4, Oct. 12, 1934, p. 339. Clark, J, G. D. s The Mesolithic Age in Northern Europe, 

prototypes, and the Hamburg culture of Meiendorf, which was ancestral 
to the Ahrensburg of Period I, was partly contemporaneous with the 
French Magdalenian. 

In the very north of Scandinavia, during at l(iast the latter part of 
Period I, tanged-point using people lived beyond the upper edge of the 
glacier, isolated, on a narrow coastal fringe, from the rest of mankind. 
To the south of the ice, in Denmark and the southern end of Sweden, as 



well as in northern Germany, Palaeolithic survivors apparently invented 
the so-called "Lyngby" antler axe to cope with the encroaching forest. 
However, the most important culture of this period was the Ahrensburg, 

Komsa Curt UPC- 


Fosna Culture 




A Swiderian 

o Hamburg 

X teemoucnampa 

MAP 1 

After Clark, J. G. D., The Mesolithic Settlement of Northwestern Europe, 1936, Fig. 15, 
p. 55; Antiquity, vol. 12, 1938, Fig. 2, p. 158. 

which is best known from recent excavations at Stellmoor, a few miles 
northeast of Hamburg. 

In Boreal times, Period II (ca. 6800-5600 B.C.) the bed of the North Sea 
rose, and England and Scotland were joined to Denmark by dry land. 
At the same time the Baltic became a fresh-water lake, called the Ancylus. 
The entire North European plain, from England over to the Urals, was 
covered with a temperate forest. In this forest arose the Maglemose cul- 
ture, which was derived from three elements: (1) the previous cultures of 


the north, especially the Ahrensburg and the "Lyngby axe" cultures, 
(2) the Magdalenian, already incorporated in the Ahrensburg, (3) the 
northward moving Tardenoisian, of eventual southern and southeastern 
inspiration. Hence, the inhabitants of northern Europe in Period I 
(ca. 11,800-6800 B.C.) were simply the descendants of the people who 
had lived immediately south of the ice-sheet for several thousand years, 
while during Period II (6800-5600 B.C.) they were joined by the vanguard 
of newcomers, whom we may already have met at Ofnet. These facts 
should give us some idea of the physical composition of the North Euro- 
pean population during these two periods. 

The culture of Period III, which roughly coincides with the time of the 
Atlantic climatic phase, was conditioned by further changes of environ- 
ment. The waters of the ocean rose due to the melting of the ice, the 
Baltic again became salt, and Britain was isolated from the continent by 
the encroachment of the North Sea. During Period III, the Maglemose 
manner of living persisted in the North European peripheries, while in 
the special center of Denmark and southern Sweden developed a new 
culture, the Erteb^lle. 

The Erteb^lle people were gatherers and eaters of molluscs on the shores 
of the newly formed Litorina Sea, and at the same time salt-water fisher- 
men. Their kitchen-middens, or shell-heaps, contain quantities of animal 
bones and discarded artefacts, including a crude pottery ware. 

The Erteb011e culture had its roots in the Maglemose, although un- 
doubtedly it was reinforced by new elements presumably from the south 
and east, and it was a local development from local human material. 
When the Neolithic farmers and herders finally reached Denmark and 
southern Sweden, they found a sedentary population of clam-diggers 
and fishermen firmly established along the coast, and owing to the 
abundance of sea food, this population must have been one of con- 
siderable density. Well-equipped sea fishermen on the shores of richly 
stocked waters offer far more resistance to invading agriculturalists 
than do hunters. The Ertebjzflle people were not driven out by their 
Neolithic neighbors, nor were they absorbed without trace into a larger 
population. In southern Scandinavia the old racial elements persisted 
alongside and in combination with the new, while in Norway the old 
tanged point makers lived on, to contribute technical methods to the 
Neolithic craftsmen. 

Period I is not represented by a single piece of human bone which can 
be dated with any pretense of accuracy. Period II, the Maglemose, is 
known from a number of skeletal finds, most of which, however, are in 
doubt. The only remains which are completely accepted and about which 
there can be no question are: (1) Stngenas, near Roe, in the Parish of 


Bro, Bohusian, Sweden; (2) Mullerp, Denmark; (3) Svaerdborg, Sweden; 
(4) Sandarna, Sweden. 34 

Of these, the only useful specimen for racial deductions of any conse- 
quence is that of St&ngenas, consisting of a brain case, a femur, and a tibia. 
These were the bones of an extremely tall man, 181 cm. in height, 35 with 
long legs, particularly in the lower segments. The femur and tibia show all 
the peculiarities of form and development associated with Upper Palaeo- 
lithic man. The brain case, which is of extreme length, has an index of 
71.9, a broad forehead, and prominent browridges. Furst, who has studied 
this fragment carefully, assigns it without question to the Upper Palaeolithic 
racial group, especially to the central European Aurignacian. 36 

The Mullerp and Svaerdborg finds consist of a child's mandible each, 87 
and a few broken fragments of other bones. That of Sandarna is limited to 
one long bone. For further evidence of the racial composition of Magle- 
mose man, we must turn to northern Germany. 

In northern Germany, which forms a part of the North European 
Mesolithic area, a number of skulls, found under varying circumstances, 
have been attributed to all three Mesolithic periods. It is difficult, if not 
impossible, to verify the alleged age of any one of them. 

Among the most likely are two adult skulls, and one adult and three in- 
fantile mandibles, dredged from the bottom of the Pritzerber Sea, north- 
west of Brandenburg on the River Havel. 38 Although they came from a 
layer of blue clay underlying peat on the lake bottom, the exact geological 
age of these formations cannot be established. Antler and bone artefacts 
recovered from the same clay belong to Periods I, II, and III of the 
Mesolithic. 39 The two skulls are probably female, although the sex has not 
been conclusively determined. Both are of dolichocephalic type, with 
indices of 71 ; both have certain early European characters such as alveolar 
prognathism, strong browridges, high temporal crests, marked supramas- 
toid ridges, and relatively large teeth. But the two differ in some respects; 
number one is large headed, short faced, and chamaerrhine, and number 
two of small capacity, very long faced, and mesorrhine. But both have 
only moderate bizygomatic diameters. 

34 Clarke, J. G. D., op. cit., pp. 133-136. 

36 Calculated by two of Pearson's formulas. Stat. 71.272 -f 1.159 X (F + T), 
and Stat. - 71.443 + 1.22F + 1.08T. 

* fl Furst, Carl M., FKVA, vol. 20, 1925, pp. 274-293. 

37 The so-called Homo kiliensis, a child's skull, may also date from this period, but the 
evidence is not sufficient for certainty. 

Clarke, op. cit., p. 133. 
Rcche, O., AFA, vol. 21, 1925, p. 176. 

Kossinna, G., MannusB, #6a, 1928; Ursprung und Verbreitung der Germanen, pp. 134 

* Reche, O., AFA, vol. 49, 1928, pp. 122-190. * 9 Clarke, op. cit., p. 134. 


A careful comparative study of these two crania places them both with- 
out difficulty into the female series of Upper Palaeolithic skulls; the great- 
est similarity is between number two and the female of Obercassel. Al- 
though the two Pritzerber crania differ widely in size and in face form, 
these differences can be matched in the Upper Palaeolithic group. The 
adult male mandible found with the Pritzerber crania is large, wide, high, 
and has everted gonial angles; it belongs to the same racial category as the 
crania. Typologically, these Pritzerber See remains are Mesolithic, for 
the two skulls could be female counterparts of Stangenas, but to which of 
the three sub-periods they belong, we cannot tell. 

The Anthropological Institute of Kiel University possesses a number of 
skulls from Schleswig-Holstein of purported Mesolithic age, most of which 
were removed from Kiel Harbor or the Kaiser Wilhelm Canal, during 
dredging operations. Others were simply dug up from peat deposits by 
farmers draining their bogs. Dates of varying accuracy may be assigned to 
seven of these specimens; all seven belong to the physical type of which 
Stangenas may be a male, and the Pritzerber See crania female, ex- 
amples. 40 

Four of them have been dated by pollen analysis; 41 three being assigned 
to the earliest Litorina transgression, and presumably to the very end of 
Period II, or Maglemose, archaeologically; and a fourth to full Litorina, 
thus probably Period III, or Erteb011e. The other three specimens, in- 
cluding the Ellerbek skull, which was dredged from submerged land in 
Kiel Harbor, may be dated, very tentatively, only by their associations 
with implements. 

Thus, so far, we have found only dolichocephalic crania of European 
Upper Palaeolithic type associated with early post-glacial Mesolithic re- 
mains in northern Germany, as well as in Scandinavia. But there are 
other skulls, of dubious Mesolithic association, which are brachycephalic. 
These include the skulls of Plau, Mecklenburg; Domitz, from the bed of 
the Elbe; and Spandau, from the mouth of the Spree. All three could fit 
very easily into the brachycephalic group from Ofnet, and if they are not 
Mesolithic, show the northward movement of that type in later times. 42 

Before concluding this survey of racial associations in the Mesolithic of 

40 Aichel, Otto, Der deutsche Mensch. The specimens referred to are B 5, KS 11032, 
KS 11254b, B 38, B 34, B 37, B 10. 

41 With newly exhumed skulls, Professor Aichel sent earth or peat from the cranial 
cavity to palaeobotanists; with specimens which had lain for years in museums, he 
gathered earth from the ear holes. This method does not always yield certain results, 
for sometimes the samples do not contain enough pollen for statistical study. 

Clarke, J. G. D., op. cit., pp. 133-136. 

Reche, AFA, vol. 49, pp. 122-190. 

Kossina, G., MannusB, #6a, 1928, p. 144. 

Kossina, G., Die fndogermanen, MannusB, #26, 1926, p. 16. 


northwestern Europe, we must not fail to mention the parallel situation in 
the British Isles. Briefly, during the Upper Palaeolithic there are no true 
Solutrean or Magdalenian deposits in Britain, but the Aurignacian con- 
tinues, to develop into an early Mesolithic culture called Greswellian. 
This in turn is later influenced by Azilian cultural diffusion from western 
France and northern Spain. The Aurignacian which came to England, 
and from which Creswellian developed, apparently came from central 
Europe. 43 

During the Mesolithic, a northern extension of Greswellian, strongly 
mixed with Azilian, extended to southwestern Scotland, where it has been 
found in the Oban caves of Argyllshire. The deposits of some of these 
caves date from Late Atlantic time, subsequent to the maximum Litorina 
transgression, during which period the caves were formed. This would 
roughly correlate the remains which they contained with Period III in 
Scandinavia. We must remember, however, that, although a few stray 
Maglemosian finds have been made in eastern Scotland, the land con- 
necting Scotland with Denmark in the Boreal period had since sunk below 
the North Sea, and skeletal material from the Oban caves cannot be 
closely related in a cultural sense to that from Scandinavia. 

During the last century, a number of these caves, when excavated, 
yielded skeletal material dating from the Late Mesolithic through the 
Bronze and Iron Ages into modern times. One of the sites, the Mac Arthur 
cave, contained some artefacts which have been recognized as Azilian, 44 
as well as two male skulls, of which one at least is probably contemporane- 
ous with the deposit. 45 

This specimen, called skull B, is very similar to the Stlingenas fragment 
in Sweden, with nearly identical vault dimensions, a cranial index of 70, a 
broad forehead, and heavy browridges. The sagittal arcs of the skull, the 
breadths and heights of the orbits, the depressed root of the nose, the 
breadth of the face, and the height of the mandible, are all typical of the 
purely long-headed variety of the Upper Palaeolithic European racial 
group. From the photographs 46 it is possible to make further observations, 
and even to reconstruct tentative values of additional measurements. The 
bizygomatic facial breadth was greater than the breadth of the vault, and 
the nose was leptorrhine. 

Oban man is, in short, an ideal example of the central European Aurig- 
nacian physical type. As far as this one specimen is concerned, the initial 

43 Garrod, Miss D. A. E., RBAA, Pres. Ad., sec. H., vol. 4, 1938, pp. 1-26, viz. p. 23. 

44 Abb6 Breuil (PSAS, vol. 55, 1921, p. 163) states that the site is fundamentally 
Creswellian influenced by a strong Azilian admixture, with faint Maglemose traces. 

46 Turner, Sir W. (TRSE, vol. 51, 1914-15, pp. 211-214), states that skull B actually 
lay in the shell deposit, while skull A was taken from the black earth above it. 
46 Turner, p. 213. 


Upper Palaeolithic invasion of the British Isles was still represented, many 
thousands of years later, by its original racial type. 


The Mesolithic Age in Europe is the time gap, lasting nine thousand 
years, between the end of the glacial period and the Neolithic. Peoples 
living in a Mesolithic stage of culture continued to obtain their food by 
hunting, fishing, and the gathering of wild vegetable products as they had 
in the Palaeolithic; but they now possessed one domestic animal, the dog. 
Technologically, the introduction of the microlithic blade, which could 
be used in composite tools, and the invention of the wood chopping axe, 
further distinguish this period. 

With the retreat of the last ice cap, the fertile grasslands of the Sahara 
and of southwestern Asia began to dry, game became scarce, and the rain 
belt moved westward and northward. The microlithic technique, which 
had been employed during late Upper Palaeolithic times in North Africa 
and the Near East, was carried across the Straits of Gibraltar and through 
the Caucasus and South Russia into Europe, where it spread northward 
and northwestward, eventually affecting the industries of the entire con- 
tinent. At the same time, it was diffused into Palestine. 

This cultural diffusion to the north and northwest was accompanied in 
the Late Mesolithic or followed in the Early Neolithic by the invasion of a 
people new to Europe, a relatively small-headed, short statured, effeminate 
looking, Mediterranean type, of direct Galley Hill deviation, and ancestral 
to one branch of the modern Mediterranean race. These Mediterraneans 
also entered Palestine in Late Mesolithic times. 

A population of Upper Palaeolithic derivation, compounded of early 
Aurignacian and Magdalenian elements, moved northward from Spain 
and the Pyrenees to western France and Germany, if not farther. In the 
vanguard of the northern movement was a large, broad-faced, brachy- 
cephalic type reminiscent of the brachycephalic element resident in Al- 
geria at an earlier date. Although its geographical origin is not certain, 
it was definitely a member of the middle and late Aurignacian group of 
mixed j^Vw^-Neanderthaloid derivation. 

In eastern Europe it is possible that the older Palaeolithic race was rein- 
forced in early post-glacial times by an increment from the Near East, 
although this cannot as yet be clearly demonstrated. Invasions parallel to 
that which crossed Gibraltar probably entered Europe via the Caucasus, 
but there is at present very little evidence to suppdrt such a theory. In 
northwestern Europe, especially in Scandinavia and Britain, where the 
last glacier had its main centers and lasted the longest, Upper Palaeolithic 
man of the central European variety persisted through the Mesolithic, and 


it is to this corner of Europe that we must look for a maximum survival of 
glacial age European man into the present time. Similarly, the other main 
type which we traced during the Palaeolithic the long-faced and long- 
legged ancestral Hamite of East Africa persisted in East Africa into the 
Mesolithic without change. 

Although most if not all of the innovations, both racial and cultural, 
which reached Europe during the Mesolithic Age, came from North 
Africa and also perhaps from points farther east, we may suppose that 
events of far greater importance to human history were going on at this 
time on the continent of Asia. There the peoples of the western plateaux 
must have already begun the mastery of the animal and vegetable worlds 
which was to permit them to increase in numbers, and to overflow into 
Europe, thus marking the beginning of the Neolithic period in the latter 
continent. If we are to understand the racial changes which affected the 
population of Europe at the end of the Mesolithic, we must next devote our 
attention to these imminent invaders. 

Chapter IV 


The word Neolithic has two meanings, one purely technical, and the 
other of broader implications: (1) the manufacture and use of polished 
stone implements, in the form of axes, adzes, gouges, chisels, and hoes; 
(2) the conquest of the procreative forces of the biological world, through 
agriculture and animal husbandry. These two definitions, implying tools 
on the one hand and food on the other, do not always overlap, for some 
peoples may be considered Neolithic in one of the two senses only. Of 
the two, only the second is of really vital importance in human history. 
In fact, the change from food-gathering to food-producing was the greatest 
step in human development since the invention of language. 1 

The initial adoption of a Neolithic economy occurred, however, at few 
centers on the earth; one in the Old World and another in the New are all 
of which we can be sure at present. In the Old World, the plants and 
animals which were suitable for domestication ranged in a wild state in 
the highland zone from Anatolia to the Indus, with some species extending 
out along the southern shore of the Mediterranean. Abyssinia may have 
been a separate center for the domestication of some grain plants, but 
probably not of animals. Perhaps when the Yemen shall have been 
studied by economic botanists, this fertile highland on the other side of 
the Red Sea will assume a like importance. 

In the millennia during which the glacier was retreating to its Scan- 
dinavian center and growing thinner, the climatic zones which made a 
well-watered grassland of this entire plateau belt moved northward, and 
the regions in which Old World civilization originated grew gradually 
drier. Afghanistan and Iran, now for the most part nearly desert plateaux, 
were then fertile; in Egypt the valley of the Nile was a string of swamps 
ind jungly lakes, full of crocodiles and hippopotami. 

It is now generally believed, although still unproven, that agriculture 
ind the domestication of animals did not arise in the three valleys of the 
^ile, Tigris-Euphrates, and Indus, but in the highlands between them. 
The river valleys became important as centers of civilization because 
icasonal flooding and the deposit of fresh alluvium made it impossible for 

1 Childc, V. Gordon, The Dawn of European Civilization; The Most Ancient East; The 
Danube in Prehistory; New Light on the Most Ancient East; Man Makes Himself. 



primitive farmers to exhaust the soil, thus permitting sedentary residence; 
furthermore, the development of irrigation and drainage canals were 
public works necessitating social solidarity, and kingdoms arose here while 
the highlanders kept to their villages and fought their feuds, as many of 
them still do today. 

As Childe has pointed out, the acquisition of a new and more productive 
means of economic life has as one of its first effects an increase in the pop- 
ulation. Agriculture and the domestication of animals did not appear in 
one day. The acquisition of a full Neolithic economy may have taken 
one or more millennia, and it only very gradually replaced hunting and 
collecting. The primitive slash-and-burn system, which must have been 
the first followed, and which was the earliest in Europe, prevents intensive 
use of the soil and promotes a slow but nevertheless positive type of 

The desiccation which followed the movement of the rain zones north- 
ward resulted initially in the migration of peoples into Palestine, North 
Africa, and southern Europe, in the form of the Mesolithic invasions, which 
we have already studied. These movements were not extensive, however, 
because the new economy of food production permitted a greater utiliza- 
tion of the drying soil on which wild animal and vegetable life, useful to 
man, had grown scarce. For a while emigration was unnecessary; but 
when the inevitable population increase had come, western Asia over- 
flowed, and farmers moved into regions where the climate which had 
formerly blessed their homelands now prevailed. 

The desiccation which followed the shifting of the cyclonic storm belts 
did not become complete until what is called, in northern Europe, Atlantic 
time, that is, in the neighborhood of 5000 B.C. Only by this time had 
Europe, south of the newly formed northern forest, really become cli- 
matically what the highland belt had been before a temperate, well- 
watered parkland, instead of a chilly, treeless plain. 

The same general date, 5000 B.C., may;be tentatively set as the time of 
the beginning of agriculture and animal domestication. It was not until 
almost 2000 years later, however, that the disciples of this new economy 
were to expand and invade more than the threshold of Europe. 

The Neolithic invaders of Europe, seeking new lands for farming and 
grazing, came as a further result of the same environmental shift which had 
impelled the earlier Mesolithic invaders, whom they supplemented without 
a gap, and with whom they blended. But the Neolithic invasion was not 
as simple as the Mesolithic. As the new economy spread, it affected a 
number of peoples, whose reactions were not all the same. Europe, the 
new stronghold of a lost climate, was broached in different places and in 
different ways. 





Map 2 will shdw, in a very general sense, the time scale of Neolithic 
invasions into Europe, and the routes by which these invasions may have 
come. It is to be noted that Crete became Neolithic before any of the 
European mainland, followed by Greece and the land near the Bospo- 
rus; eventually these agriculturalists spread into all the northern Mediter- 
ranean lands by sea. Meanwhile, other Neolithic farmers had been moving 
along the coast of North Africa from Egypt, and had crossed over Gibraltar 
to invade Spain. Hence they migrated northward and eastward, as far 
as the Swiss lakes and the Rhine. 2 Their agriculture, and their pig, sheep, 
and cattle husbandry, eventually spread over most of western Europe, 
and even into England. At the same time still other farmers, in this case 
coming from Anatolia, or southeastern Russia, or both, were moving up 
the Danube, and eventually established themselves in the fertile valleys 
of Moravia and Bohemia, and even farther westward until they met the 
stream coming northward over Gibraltar. 

These three movements were the primary invasions which brought a 
new, agricultural population into Europe. Later in the Neolithic there 
were two other movements of a different character. One was that of the 
Megalith-builders, who sailed through the Straits of Gibraltar and skirted 
the western shores of Europe to the British Isles and Scandinavia. These 
seafarers probably introduced the new economy to the northern isles and 
Scandinavia. Then there were the Corded people, so-called on account 
of the decoration on their pottery who came from some mysterious point 
in southern Russia or the steppes of western Asia north of the plateau, 
and who were probably less dependent on farming than on pastoral 
nomadism and trade. Just as the Megalithic people carried civilization 
to the far western corners of Europe by sea, so the Corded people intro- 
duced the new enlightenment into the north, where the old hunting and 
fishing life survived. 

Five invasions, then, converging on Europe from the south and east, 
brought a new population to Europe during the third millennium B.C., 
and furnished the racial material from which living European populations 
are to a large extent descended. 


In Europe, the Neolithic is primarily the period of the Mediterranean 
race, in one form or another. It was, apparently, the Mediterraneans 
who accomplished the change to a food-producing economy elsewhere, 
and who expanded into the territory of the food-gatherers. 

These Mediterraneans, while surprisingly homogeneous in some re- 
spects, may be segregated locally and typologically into sub-groups on 
8 Menghin, O., Weltgeschkhte der Steinzeit, pp. 294-302. 


the basis of a few characters. Before proceeding much further with our 
geographical-historical reconstruction, it will be well to define what we 
mean by Mediterranean, to compare it with other races which we have 
already met, and to specify its principal subdivisions. 

By Mediterranean, in the skeletal sense alone, we mean the wide family 
of closely knit racial types which are long headed, orthognathous, meso- 
rrhine or leptorrhine, narrow faced, and of medium head size, descended 
from the general Galley Hill stock, and related to Combe Capelle and 
Afalou #28. Mediterranean, in this sense, is the name by which we pro- 
pose to designate that one of the two major racial elements, concerned with 
the development of white peoples, which completely lacks Neanderthaloid 
ancestry. It differs from the major Upper Palaeolithic group of Europe 
and northern Africa in several respects, as shown on page 84. 

The "Mediterranean" racial family is just as " white," in the larger 
meaning of the word, as the Upper Palaeolithic family. Its chief differ- 
ences from the latter are: a smaller brain size, a moderate body size, and 
a lack of the excessive specializations which characterize the northern 
group. The Mediterranean group seems to be of purely sapiens ancestry, 
without Neanderthaloid or other mixture. 

Before the Neolithic, the principal branches of the Mediterranean 
family must already have come into existence. Some Mediterraneans 
were probably white skinned, and others brown; it is also possible that the 
differences in hair and eye color which so strongly distinguish living 
Mediterranean sub-varieties had already come into existence. 

We cannot speak with authority about Nordics until we meet blondism 
in the flesh, nor make profitable surmises about them until we find it in 
literary references and artistic representations. We must not, therefore, 
let differences in pigmentation and soft parts confuse our understanding 
of the skeletal unity of the Mediterranean race. 

It can be shown that Sumerians who lived over five thousand years ago 
in Mesopotamia are almost identical in skull and face form with living 
Englishmen, and that predynastic Egyptian skulls can be matched both 
in a seventeenth century London plague pit, and in Neolithic cist-graves 
in Switzerland. Modern dolichocephalic whites or browns are very similar 
in head and face measurements and form. The Nordic race in the strict 
sense is merely a pigment phase of the Mediterranean. 8 

On the basis of the material to be covered in this chapter, we may dis- 
tinguish the following branches of the general Mediterranean or Galley 
Hill group: 

1 Popularly, the word "Nordic" is frequently applied to a blond or pigmentally inter- 
mediate conglomerate type or group of types in northern Europe, which contains other 
than blond Mediterranean elements. 



Upper Palaeolithic 

1 . Great size of brain case. 

2. Mean skull length about 198 mm. 
in males. 

3. Vault height variable, usually mod- 

4. Head form variable. Local means 
70-72 in some cases, 74-75 in others. 

5. Strong tendency to become brachy- 
cephalic, manifest in some local 

6. Thick vault, heavy relief of muscle 

7. Browridges and development of 
nuchal lines on occiput strong. 

8. Face variable in length, frequently 

9. Face very broad, bizygomatic di- 
ameter over 140 mm. in males. 
Zygomatic arches greatly bowed. 

10. Orbits very wide and low. 

11. Distance between orbits great. 

12. Nasal skeleton prominent. 

13. Sub-nasal segment of face height 
relatively great. 

14. Mandible thick, heavy, with great 
symphysial height, wide bicondylar 
and bigonial diameters; prominent, 
often bilateral chin. 

15. Stature variable, but most char- 
acteristically tali, mean probably 
about 172 cm. 

16. Bodily build usually robust, shoul- 
ders very broad, chests voluminous, 
hands and feet large. 


Brain size variable, but usually moder- 

Mean skull lengths between 1 83-1 93 mm. 
in males. 

Absolute vault height has the same 
range in absolute dimensions, or higher, 
but usually higher in relation to other 
diameters. Within the Mediterranean- 
Galley Hill group, differences of vault 
height serve as diagnostics of race or 

Tendency to brachycephaly not mani- 
fested by advent of Neolithic in areas yet 

Vault medium to thin, muscular relief 
on vault as a rule slight. 
Browridges and nuchal lines variable, 
medium to weak. 

Same, but some notably long-faced 

Face usually narrow, 127-133 mm., 
zygoma tic arches weak and laterally 

Orbits of moderate proportions. 
Distance between orbits slight. 
Nasal skeleton prominent in some types, 
but not in all. 

Sub-nasal segment of face height rela- 
tively slight. 

Mandible variable; usually light, of 
small symphysial height, and narrow in 
both lateral diameters, chin moderate 
or pointed. In some types, however, the 
mandible approaches the Upper Palaeo- 
lithic form in height, but not in breadth. 
Stature variable, but most characteris- 
tically short, types vary from 159 to 172 
cm. in means. 

Bodily build usually linear, hands and 
feet smaller, weight probably less. 

(1) Mediterranean Proper (hereafter meant when the word "Mediter- 
anean" is used alone): Short stature, about 160 cm.; skull length 183- 
187 mm. male mean; vault height 132-137 mm. mean; cranial in- 
lex means 73-75; browridges and bone development weak, face short, 


nose Jeptorrhine to mesorrhine. Type already met in Portugal and 
Palestine in Late Mesolithic. Represents the paedomorphic or sexually 
undifferentiated Mediterranean form, and often carries a slight negroid 

(2) Danubian: The same in body size and build, skull length and cranial 
index the same; individually, the index goes to 80. Vault is higher than 
breadth, means 137-140 mm. Nose mesorrhine or chamaerrhine. 

(3) Megalithic: Tall stature, means 167-171 cm., slender build; skull 
length over 190 mm.; cranial index 68-72 means, individual range below 
78; vault moderate in height, less than breadth; forehead moderately 
sloping, browridges often of moderate heaviness, muscular markings 
stronger, skull base wider, face medium to long, nose leptorrhine, mandible 
often deep and moderately wide. The East African Elmenteitans repre- 
sent an individual and extreme form of this. It represents a gerontomor- 
phic or sexually differentiated Mediterranean or Galley Hill form, and 
in cranial features is closer to Galley Hill itself than any other branch. 

(4) Corded: Tall stature, means 1 67-1 74 cm. ; build linear but muscular, 
perhaps heavier than the Megalithic; extremely long-headed, 194 mm. 
mean. Vault of great height, means over 140 mm., exceeding breadth; 
browridges and muscular markings medium to strong; face very long, and 
of slight to moderate breadth; mandible deep and chin marked, but nar- 
row through gonial angles. Nose leptorrhine, often prominent. This type, 
in western and northern Europe, approaches in some respects the Upper 
Palaeolithic type with which it mixed. 

(5) Other Forms: Include mixtures between the four named, as well as 
others which are also intermediate but perhaps ancestrally undifferen- 
tiated. The later "Nordic" forms are intermediate. In Asia Minor and 
the Irano-Afghan plateau appear forms noted for great prominence and 
convexity of the nasal skeleton, and lack of nasion depression. Since thqse 
features are found on individuals of varying size and proportions, as well 
as brachycephalic races of the same neighborhood, they seem to represent 
some local genetic tendency, and cannot be considered the exclusive prop- 
erty of a given race. However, one might name the small variety found in 
Asia Minor Cappadocian, while a larger form commoner farther east, and 
metrically close to the Corded, may be called Afghanian. 

The names given the racial divisions outlined above have been chosen 
with the intention of avoiding close reference to living races, since they are 
based on the skeleton alone. Mediterranean forms an exception; it is so well 
known and firmly established that it cannot be changed. In this particular 
case, we may be reasonably sure of the character of the soft parts, owing to 
the antiquity of accurate realistic portraiture in Egypt, Crete, and Meso- 
potamia, as well as to mummification. 


The names Danubian, Megalithic, and Corded, have been deliberately 
taken from archaeology since, as will be shown, the types so designated 
were closely linked, during the Neolithic and even later, to the cultural 
entities with which they are thus identified. 

It is hoped that the use of these labels will eliminate the necessity, in 
the rest of this chapter, of elaborate description. 


Unfortunately for the compiler of a general book, both the archaeology 
and the somatdlogy of the Iranian plateau and of Mesopotamia are in 
their respective infancies. City after city, and village after village, remain 
undug, while thousands of skulls, some excavated and discarded, and 
others still in the ground, remain unmeasured and unpublished. Despite 
the notable work done at al ? Ubaid, Kish, Ur, Warka, Susa, Persepolis, 
Rayy, and other sites, the archaeologists have not yet found the begin- 
nings of Near Eastern civilization. Until recently, no single unquestion- 
ably Neolithic site had been discovered in the whole of Asia Minor, Meso- 
potamia, or the Iranian plateau over to India; at present several sites 
have been located in Anatolia and Armenia. 4 In the Tigris-Euphrates 
Valley, the Neolithic material, if it exists, must be buried by many feet of 
alluvial soil. In the eastern highlands, if it exists, it should not be hard to 
find. The difficulty is that no one has seriously looked for it. The claim 
of this whole highland and riverine zone to priority in the development 
of the Neolithic economy cannot yet be confirmed or refuted. 

There is, however, another claimant equally lacking in credentials the 
plain of west-central Asia, north of the plateau, and east of the Caspian. 
In the grasslands drained by the Oxus and Jaxartes the great nomadic cul- 
tures, associated with Indo-Aryan-speakers in the oldest traditional times, 
and later with the Turks, had their bed of germination. From this center, 
from time to time, invasions and migrations started in several directions. 
One was the movement of the Aryan ancestors into India, about 1400 
B.C.; another the Iranian invasion of the plateau which bears the name 
and whose inhabitants speak the language of the invaders. A school, 
founded by the Indo-European philologists of the last century, and sup- 
ported, although with different dramatis personae, by the modern Turks, 
would make of these vast plains the germinating-bed of Old World food 

Commencing with the Iranian plateau, we may consider skeletal ma- 
terial which antedates the arrival of the Iranian-speaking immigrants. 
Five crania from Luristan and the region directly to the north, in western 

* Pittard, E., ASAG, vol. 7, 1937, pp. 389-391. 
Field, H., AJSL, vol. 55, 1938, pp. 101-111. 


Iran, represent the pre- Aryan period; three males and two females, 6 dat- 
ing from 2000-1100 B.C., are all variants of the general Mediterranean 
type, in Cappadocian and Afghanian directions. An Early Copper Age 
skull from southern Baluchistan, which may date from the third millennium 
B.C., is the same. 6 We may surmise that the ancestors of the bulk of the 
present plateau population had arrived by the beginning of the third 

In Mesopotamia, the earliest cultural remains have been found in 
Sumeria. Here there has been recognized a long predynastic period, sub- 
divided into three phases al 'Ubaid, Uruk, and Jemdet Nasr. These 
three probably occupied the fourth millennium B.C. The last two at least 
were Copper Age cultures, while the al TJbaid culture proper, as exempli- 
fied by the eighteenth to fourteenth levels at Warka, may possibly have 
its roots in a true Neolithic. 7 

One grave at Warka, in level 14, belongs to the latter part of the 
al 'Ubaid period, probably about 3700 B.C. The skull contained in it is 
said to be dolichocephalic. Two skeletons from perhaps equally early 
graves at al 'Ubaid itself powdered upon exposure, and could not be 
measured. Hence our knowledge of the people of the fourth millennium 
B.C. in Mesopotamia, based on indubitably contemporaneous remains, is 
practically zero. 

A series of seventeen crania from al 'Ubaid 8 (see Appendix I, col. 4), which 
may be predynastic or early dynastic, belong without exception to a type 
which has been called Eurafrican, and which has been the most numerous 
and most characteristic element in the population of Mesopotamia from the 
time of the marsh dwellers at al c Ubaid to the present day. These skulls are 
large, heavy, and purely dolichocephalic. They belong to the larger- and 
longer-headed Mediterranean division, nearest in vault size and form to 
the earlier Galley Hill and Combe Capelle. They differ in one important 
respect, however, from most European skulls of the same general type, 
in that their nasal bones are extremely prominent and highly placed. 
These early Sumerians, like the inhabitants of the Iranian plateau, had 
already acquired the projecting, aquiline noses so characteristic of the 
modern Near East. Like the plateau dwellers, these early Sumerians 
were Afghanian in race. 

Mesopotamia is not, like Egypt, an isolated valley, for it may be entered 

'Vallois, H. V., "Notes sur les Tfctes Osseuses," in Contencau, G., and Ghirsh- 
man, A., Fouilles de Tepe Giyan. 

6 Sewell, R., and Guha, B., Report on the Bones Excavated at Nal, MASI, vol. 35, 1929, 
app. 5, p. 56. 

'Jordan, J., APAW, Jh. 1932, #2. 

8 Keith, Sir Arthur, "Report on the Human Remains, Ur Excavations," vol. 1 : in 
Hall, H. R. H., and Woolley, C. L., Al Vbaid, 


without great difficulty from the highlands to the east and north, while it 
forms a natural goal for the inhabitants of the Arabian uplands, made 
mobile by the fickle rainfall of the pastures. The history of Mesopotamia 
has consequently been a sequence of infiltrations and invasions from both 
the highland zone and the deserts, for the country feeds the city with 
men, and not the reverse. 

In studying the racial history of Mesopotamia from the third millennium 
B.C. onward, we must remember this almost constant influx, and observe 
how it affected the Sumerians and the Semitic-speaking kingdoms. The 
series of skeletal remains at our disposal, other than the series from al 
'Ubaid, include: (a) a series from Kish, from graves which may be dated 
at some time close to 2900 B.C.; (b) another from the same site, from 
fourth dynasty graves, prior to 2500 B.C.; (c) skulls of the third dynasty 
of Ur, dated about 2300 B.C. (d) Neo-Babylonian crania, from between 
800 and 400 B.C. (e) Skulls from Kirkuk dated at the fifth century A.D. 9 

In all, well over a hundred skeletons have been studied. Most of the 
skulls belong to the "Eurafrican" type already described, but two other 
types are represented in most of the series. One of these is an ordinary 
Mediterranean with a smaller skull and a higher cephalic index, which 
ranges between 70-80 and averages about 75. This Mediterranean type 
is more fragile, less rugged, shorter faced, and smaller in body size. This 
is apparently not an original Sumerian type, for it is completely absent 
in the earliest series from al 'Ubaid and Kish. It first appears well after 
3000 and probably after 2700 B.C. in the fourth dynasty graves at Kish 
(see Appendix I, col. 5), and from then on seems to persist in all of the 
samples, except for the late Kirkuk series in the north. Like the larger 
"Eurafrican" this smaller Mediterranean type may still be distinguished in 
the living population of Iraq. . 

The "Armenoid" racial type, which is the third one claimed in Meso- 
potamia, begins with the earliest Kish graves and continues through the 
Babylonian period. The identification of this type is not wholly certain, 
however, for very few actually brachycephalic skulls have been found, 
and, since facial portions of these are usually damaged, it is impossible to 
define the type clearly. Most of the so-called Armenoid skulls are meso- 
cephaiic or sub-brachycephalic, but, in a few instances, the cephalic 
index. runs really high, in an extreme case, to 89. The occiputs of these 
skulls are said to be flat, the browridges heavy, and the capacities great. 
Although many of the skulls which have been called Armenoid may rep- 

8 (a) Penniman, T. K., in Watelin, L. C., Excavations at Ktsh t vol. 4. (b) Buxton, 
L. D., in Langdon, Excavations at Kish, vol. 1 ; also Buxton, L. D., and Rice, D. T., JRAI, 
vol. 61. (c) Keith, Sir A, "Report on the Galilee Skull." (d) Buxton, L. D., vide supra. 
Buxton and Rice, vide supra, (e) Ehrich, R. F., Appendix to Starr, Richard, F. S., 
vol. 1. 



resent merely the rounder headed extreme of the total group, it is never- 
theless probable that a planoccipital brachycephalic strain actually pene- 
trated Mesopotamia during the third millennium B.C. Although it has 
since increased in numbers, it still forms but a minority. 

Except for these few brachycephals, none of the invasions or cultural 
movements into Mesopotamia in historic times has changed the popula- 
tion in any perceptible way. This would indicate that the regions which 


FIG. 17 

FIG. 18 

FIG. 19 FIG. 20 

Fig. 17, Frankfort, H., Jacobsen, T., and Preusscr, C., COIC, #13, 1930/31, p. 70, 
Fig. 27. Fig. 18, King Gudea, after Woolley, C. L., The Development of Sumerian Art. 
London, 1935, Fig. 62a, p. 1 15. Fig. 19, from excavations at Khafaje, Expedition of the 
Univ. of Pennsylvania Museum and of the American School of Oriental Research, 
under Dr. E. A. Speiser, New York Sunday Times, Section 9, June 27, 1937. Fig. 20, same 
source as Fig. 19. 



acted as feeders of immigrants to Mesopotamia were themselves similar 
racially. The plateau people of Iran, therefore, were probably in the main 
long-headed. The inhabitants of northern Arabia who had entered the 
valley from time to time, and who still come to the banks of the Euphrates 
to water their flocks, belong likewise to the general Mediterranean family, 
and examples of both Afghanian and Mediterranean types may be selected 
from the living tribes without difficulty. It is quite possible that the first 
appearance of the finer and smaller Mediterranean type in Mesopotamia 
came with the arrival or assimilation of the Semites. 

The Sumerian sculptors have left behind them records in stone which 
may piece out the evidence of the skulls. These records consist of bas re- 
liefs, which are of conventional type, 
and some really excellent portraits in 
the round. The reliefs show wiry, ath- 
letic men with large, often aquiline 
noses. They are obviously normal 
white men of some Near Eastern va- 
riety, just as one would expect. The 
portrait busts, of which three examples 
are shown (Figs. 17, 18, and 19), 
seem really to depict individual men 
rather than conventional types or 
ideals. Figure 18 represents the oft- 
sculpted King Gudea, who has a 
roundish face and a nose less promi- 
nent than the bas-relief ideal. Figure 
17, which looks less posed, bears the 
sly expression of a Baghdad shop- 
keeper of the present day. In both 
heads the browridges are absent, and 
the eyebrows concurrent. In these as in most examples of Sumerian 
sculpture, there is no evidence of hair distribution or hair form which 
is, however, conventionally shown in archaic statues of gods (Fig. 20), 
dating from early dynastic times. In these, the beards are full, the hair 
straight or wavy. 10 

In the later Babylonian and Assyrian sculptures, which depict Semitic- 
speaking populations, we find a profusion of beard and head hair as with 
the early Sumerian gods; the hair is wavy or curly, and the beard ex- 
aggeratedly abundant. (See Fig. 21 .) The eyebrows meet over the root of 

10 Frankfort, H., ''Oriental Institute Discoveries in Iraq, 1933-34," Fourth Prelim- 
inary Report, COIC #19, 1935, 

Speiser, E., New Tork Times, Jan. 27, 1937. 


After Schafer, H., and Andrae, W., Die 
Kunst des alien Orients, 1925, p. 521. 


the highly arched nose, and eyelids and lips are full. The bodies are con- 
ventionally thick-set, the arms and legs heavily muscled. The artists of 
Babylon and Nineveh were anthropologists at heart, for they chose a truth- 
ful rather than an imaginary ideal. Their kings and soldiers and slaves 
could step down from the walls and mingle with the crowds today. 

Although Mesopotamia was one of the great centers of Old World 
civilization, and although its emissaries travelled hundreds of miles, and 
its cultural influences were so far-reaching that we may feel them even 
today, we must not attempt to link it directly with the Neolithic invasions 
which entered Europe. The farmers who sought rich fields and grassy 
meadows to the west of the Euxine and the Bosporus were not Sumerians 
or Babylonians, but peoples who had started their wanderings before the 
development of a metal age civilization, and who were affected only in- 
directly by cultural emanations from its center. Nevertheless, this somato- 
logical survey of early Iran and Iraq is of value in the larger problem of 
the white race, for it enables us to define clearly the physical charac- 
teristics of the Mediterranean types of man which were responsible for 
what may have been the world's earliest civilization, and of the surround- 
ing regions from which it was fed, just as one could tell the physical types 
of France from a study of Paris or of Europe from a study of New York. 


Certainly the most satisfactory area in the whole world for the racial 
study of a people of antiquity is the valley of the Nile. Over four thousand 
Egyptian skeletons, covering a period of some seven thousand years, have 
received anthropometric attention. One Egyptian series, consisting of nine 
hundred males, is the most extensive group of crania of a single sex and 
from a single place ever assembled. It is possible, therefore, to study varia- 
bility and change in this isolated valley with delicate precision, for in one 
district, the region of Upper Egypt about Abydos and Thebes, the cranial 
material is more abundant than that of any age from any other region of 
the same size in the world. 11 

Furthermore, from the beginning of dynastic times until the arrival of 
Islam, Egyptian painters and sculptors recorded faithfully, often in colors, 
the physical appearance of their living countrymen, as well as of many 
different kinds of foreigners. At the same time, the climate of the Nile 
Valley, and the skill of embalmers, have preserved intact the hair, skin, 
and dried muscles of both natural and artificial mummies, from the pre- 
dynastic period onward. With this abundance of evidence, we should be 
nearly as familiar with the racial characteristics of the ancient Egyptians 
as with those of the people of our own day. 

Morant, G. M, Biometrika, 1925, p. 4. 


Geographically, Egypt is not unified. In the first place, the Delta, which 
resembles early Sumeria in its climatic conditions, is a marshy series of 
water ways, continuous with the coasts of Palestine and Lybia, and easily 
attainable from both directions, as well as from the sea. It, and to a lesser 
extent, Lower Egypt, as well, forms an easy route of passage from Asia to 
North Africa without touching most of Egypt proper. It is possible, there- 
fore, that even in dynastic times movements of racial importance passed 
from western Asia to North Africa over this coastal route, without affecting 
the population of Egypt in any notable way. Upper Egypt, on the other 
hand, is a narrow valley hemmed in by cliffs on either side. Beyond these 
cliffs lie plateaux, which during the pluvial periods were well-watered and 
covered with grass and game. There was only one gateway to Egypt from 
the south down the Nile and during the dynastic period the Egyptian 
kings kept garrisons on their southern boundaries to prevent immigration 
from this quarter. 

The cyclonic rain belt which moved northward from the Saharan and 
Arabian deserts in the general post-glacial readjustment of climate also 
took a westward direction. 12 For this reason, a climate favorable for hunt- 
ers and gatherers persisted longer in Egypt than in Mesopotamia. At 
the same time, this movement may have had much to do with the migra- 
tion of peoples crossing North Africa from east to west, keeping ahead of 
the zone of serious desiccation. Morocco was the last of North Africa to 
dry, and in parts of that country cedar forests and grassy uplands still 

The archaeological sequence in Egypt, which has been well worked out, 
begins with the lowest Palaeolithic and continues without a gap until his- 
torical times. During the pluvial and early post- pluvial periods, however, 
the swampy tree-fringed valley was not the most favorable hunting ground, 
and Palaeolithic and Mesolithic food-gatherers ranged by preference over 
the open grasslands to either side, making only occasional visits to the 
river banks. As the plateaux grew increasingly arid, many of the hunters 
who did not migrate westward moved into the still moist valley, toward 
which the game upon which they lived must have been converging. One 
such concentration of food- gatherers is seen in the Sebilian culture of Up- 
per Egypt. 13 The skeletal remains from this culture, which have not yet 
been published, are said to anticipate in physical type the predynastic, 
placing a fine Mediterranean type in pre-Neolithic times. 14 

In another part of Upper Egypt, the earliest known of the sporadic 
agriculturalists, who at the same time or soon afterward, began to exploit 

12 This summary of climatic changes in Egypt is based on Childe, V. G., New Light 
on the Most Ancient East, pp. 49-51. 

18 Childe, op. V., p. 35. " Leakey, L. S. B., Stone Age Africa, pp. 177-178. 


the favorable environment of the Nile Valley, were the so-called Tasians, 
named after the type site of their culture at Deir Tasa. At the time of their 
occupation, this part of the Nile Valley was still swampy, with large trees 
growing at the fringes of the marsh. In view of these climatic conditions, 
it is estimated that this culture may have been introduced as early as 5000 
or even 6000 B.C. 15 

Although the physical type of the Tasians has not yet been fully de- 
scribed, Brunton's preliminary notice informs us that the few skulls as yet 
found are large, thick-walled, and strong in muscle relief, with heavy 
browridges. The cranial form, while prevailingly dolichocephalic, in- 
cludes some brachycephals. 16 The faces are broad, the orbits square, the 
lower jaws deep, wide, and square, with flaring gonial angles and pro- 
jecting, bilateral chins. Judging from the drawings of one example pub- 
lished by Brunton, we may deduce that they were orthognathous, and in 
this case at least, mesorrhine. They seem to belong to a purely white cate- 
gory, and we may hazard a guess that they represent an Upper Palaeo- 
lithic strain of Afalou or Early Natufiian type, forming a link between 
Algeria and Palestine. They were not, however, important in the ultimate 
formation of the Egyptian people, for in subsequent times they seem, both 
culturally and racially, to have disappeared. 

Another early Neolithic civilization of Egypt which left no clear traces 
in the dynastic culture was that of the Fayum people and the Merimdians 
of the Delta, who, contemporaneously with the Tasians, and following the 
Sebilians, grew barley, emmer wheat, and flax along the shores of the 
Fayum Lake and the estuaries of the Delta. They also kept herds of cattle, 
and especially of swine. Their technology bridges the gap between a Cap- 
sian Mesolithic and a full Neolithic. Their pottery, a thick black ware 
decorated by incision, resembles early ceramic types of Neolithic western 
Europe and of Anatolia. 

The importance of these people is that they probably represent the pro- 
totype of the Neolithic agriculturalists who moved westward along the 
shore of North Africa to Morocco, and over into Spain, whence they spread 
the Neolithic economy, with emmer, flax, and swine, to the Swiss lakes 
and to the Rhine. 17 Although they may have had little importance for 
Egypt, they had much for Europe. Their appearance in the Fayum and 
the Delta is dated at about 5000 B.C., and their disappearance about 4000 
B.C. One millennium later they or people like them appeared in western 

The skulls of these people, which consist mostly of females and infants, 

15 Brunton, Guy, Antiquity, vol. 3, #12, Dec., 1929, pp. 456-457. 

16 Menghin, O., Lecture at Harvard University, April 6, 1937. 

17 Childe, V. G., op. /., p. 64. 


are all dolichocephalic and Mediterranean. There is no trace of negroid 
influence, and the skulls are said to be larger than those of predynastic 
Egyptians, to be described shortly. 18 

After this excursion let us return to Upper Egypt, to a number of sites 
close to that section of the valley in which the Tasians had previously lived, 
From the type site, Badari, come the earliest skulls of a definitely Egyptian 
group which have yet been discovered. These Badarians lived about 
4000 B.C., after the climate had become considerably drier than it was in 
Tasian times, so dry, in fact, that in many cases the skin and hair of their 
dead have been naturally preserved. The skin was apparently brunet 
white, while the hair was black or dark brown in color, thick, of fine 
texture, and usually wavy in form. 

Although the Badarians, like the Tasians and Merimdians, still hunted 
and fished to enhance their larders and vary their diet, they lived primarily 
by agriculture and by herding cattle and sheep. Unlike the Merimdians, 
they raised no pigs. By hammering copper they were entering the transi- 
tion from the Neolithic to the Metal Age. They navigated the Nile in ships, 
whose shapes are revealed by pottery models, but we cannot be sure that 
they sailed them. These Badarians were undoubtedly newcomers to Upper 
Egypt, who displaced the Tasians and perhaps other predecessors. 

It is very difficult to identify the sexes of Badarian skulls, for the type is 
a delicate and feminine one, showing very little muscular development. 19 
For this reason, the various investigators who have measured Badarian 
skulls have in no two cases agreed on their sexing, and the means vary 
accordingly, but with the most extreme division, the sex ratios are still 
unusually small, even for an Egyptian series. 

The Badarian series is the earliest cranial sample of any numerical length 
which has yet been obtained from any part of the world. It is our first 
series, unified in time and place, which is ample enough to be studied by 
accurate statistical methods. These show that the series is not very vari- 
able, but its variability is no less than that of many modern populations. 
From this Morant concludes "In the last six thousand years there appears 
to have been little change in the variability of racial populations." 

The Badarian type represents a small branch of the Mediterranean 
racial group. The head is unusually high in comparison to the other di- 
mensions, and the facial skeleton is in the absolute scale unusually small; 
the mandible is small, narrow, and light. Its mean male bicondylar diam- 
eter is the smallest known, while the bigonial diameter of 91.6 mm. is also 
extremely low. 

Although the Badarian type is definitely related to that of the succeed- 

18 Deny, Douglas, SAWV, Jahrgang, 1932, #1-4, pp. 60-61. 20 Ibid., p. 306. 

19 Morant, G. M., Biometrika, 1927, vol. 27, pp. 293-309. 


ing predynastic people, it is distinguished from it in a number of ways. 
The Badarian skulls are more prognathous than those of their successors, 
and have higher nasal indices. The nasal index is just on the line between 
mesorrhiny and chamaerrhiny. In fact, while the prognathism and nose 
form would suggest a negroid tendency, this cannot be established, since 
the hair form is definitely not negroid. 

Morant shows that the Badarian cranial type is closely similar to that of 
some of the modern Christians of northern Ethiopia who incidentally 
do not show negroid characteristics in the skull and also to the crania of 
Dravidian-speaking peoples of southern India. One might add that living 
Somalis show a close approximation to this physical type in most respects, 
and the extremely narrow jaw in which the Badarians seem to reach a 
world extreme may be duplicated among both Somalis and the inhabi- 
tants of southern India. In Europe, the closest parallel to the Badarian 
type is found among modern Sardinians, but this is not as close as their 
relationships to other and later Egyptians. 

On the basis of these racial comparisons, it seems reasonable to suggest 
that this Badarian physical type may have come from the south, near the 
headwaters of the Blue Nile. It may represent an early Hamitic racial 
strain, which persists despite some negroid admixture in Ethiopia and 
Somaliland to the present day. 

The Badarian was succeeded in Upper Egypt by a sequence of cultures 
which may be treated under the collective term predynastic. In predynas- 
tic Egyptian times, the inhabitants of Lower Egypt, that is the region 
around Memphis and the modern Cairo, were physically and culturally 
distinct from those of Upper Egypt. The Egyptian writing was developed 
in Lower Egypt where reeds, birds, and other natural objects typical of 
that environment were incorporated into the syllabic and alphabetical 
signs. In predynastic times, there were two kingdoms of Lower and of 
Upper Egypt. The union of the two under Menes, around 3000 B.C., 
marks the beginning of the dynastic tradition. Predynastic times may be 
considered, therefore, to have occupied most of the preceding millennium. 

In Upper Egypt, the early predynastic physical type is best represented 
by the series from Naqada. 21 (See Appendix I, col. 6.) The Naqada peo- 
ple, although they resembled the Badarians in many respects, yet differed 
from them sufficiently in others to assure us that these were two popula- 
tions of separate though related origins. The Naqada people were fairly 
tall, with a mean stature of 167.5 cm. for eighty males. They were prob- 
ably taller than the Badarians, although we have no definite data on Bada- 
rian stature. Both heads and faces were wider and larger than those of 
the Badarians; the noses were narrower, and there was less prognathism. 

21 Morant, G. M., Biomctrika, vol. 17, 1925, pp. 1-52. 


The less numerous Badarians were probably absorbed into the Naqada 
population, though there is no direct evidence to confirm this assumption. 

In Lower Egypt lived another group of Mediterranean predynastic 
people who differed from the Upper Egyptians in certain noticeable ways. 
The heads were broader, the cranial indices higher, reaching a mean of 
75, whereas the Upper Egyptian mean is nearly 72. The vault height is 
less, the face is no broader, but somewhat longer, and the nasal index is 

The two types from Upper and Lower Egypt represent the extremes of 
a purely native Egyptian population, but from the beginning of dynastic 
times, around 3000 B.C. until Ptolemaic times, the numerous series which 
give an excellent picture of the progress of racial continuity and change in 
Egypt show the interactions of these two types. The racial history of Egypt 
in the course of three thousand years was simply the gradual replacement 
of the Upper Egyptian type by that of Lower Egypt. 22 (See Appendix I, 
cols. 7, 8.) As one looks at the tables from century to century, one sees that 
the crania increased gradually in breadth from 131 to 139 mm., and the 
faces from 124 to 129 mm. Ancient Egypt must remain the most outstand- 
ing example yet known in the world of an important, naturally isolated 
region in which native racial types were permitted to develop their own way 
for several thousand years completely uninfluenced by foreign contacts. 

Modern Copts, who probably represent the ancient Egyptian type more 
faithfully than the Moslem population, have diverged from the earlier 
types only in a reduction of the skull length from about 183 mm. to 
177 mm. Therefore, evolutionary change in Egypt consisted entirely of a 
slight reduction of head length, and in places of a lengthening of the face, 
and a narrowing of the nose; but the change has not been notable. 
Changes in physical type in any part of Europe within the last five hundred 
years have been much greater than in Egypt during five thousand. 

The wealth of contemporary illustrative material from Egyptian art 
sources may be divided into two classes, conventional representations and 
portraits. The former show a definite and well-recognized type; slender- 
bodied and wiry, with narrow hips and small hands and feet. The head 
and face are those of a smoothly contoured fine Mediterranean form. 

The portraits, on the other hand, show two things in particular: that 
there -was considerable individual variation in bodily build as in head and 
face form within the dolichocephalic and mesocephalic range, and that 
many of the officials, courtiers, and priests, representing the upper class 
of Egyptian society but not the royalty, looked strikingly like modern 
Europeans, especially long-headed ones. This is due perhaps to the fact 
that the Egyptian nose was not typically high rooted, like those of the 

22 Morant, op. at., 1925. 




FIG. 22 

FIG. 23 

FIG. 24 FIG. 25 

Fig, 22, portrait head of a man, in green slate, in Agyptisches Museum, Berlin; after 
anon., The Art of Ancient Egypt, Phaidon Press, Vienna, 1936, Plate 166. Fig. 23, portrait 
of Rahotep, Cairo Museum, after Schafer, H., and Andrae, W., Die Kunst des alten Orients > 
1925, p. 222. Fig. 24, plaster mask from workshop of the sculptor Thutmosis, Statlisches 
Museum, Berlin. Schafer and Andrae, p. 339. Fig. 25, statue in possession of the Earl 
of Carnavon, Frontispiece, Journal of Egyptian Archaeology > vol. 7, 1917. 


Mesopotamians as depicted in their art; and also, perhaps, because the 
portraiture, at least of the men, shows a greater angularity of line and form 
than do the conventional representations. 

There may also have been some distinction of type in the royal families, 
for the rulers often have that extremely dolichocephalic head form, 
coupled with a sloping forehead and high nasal aquilinity, with highly 
excavated nostrils, seen so typically in the familiar mummy of Rameses III, 
as in the living emperor of Ethiopia, Hailie Selassie. This strain may well 
have been derived in most ancient times from the headwaters of the Nile. 

The pigmentation of the Egyptians was usually a brunet white; in the 
conventional figures the men are represented as red, the women often as 
lighter, and even white. Although the hair is almost inevitably black or 
dark brown, and the eyes brown, Queen Hetep-Heres II, of the Fourth 
Dynasty, the daughter of Cheops, the builder of the great pyramid, is 
shown in the colored bas reliefs of her tomb to have been a definite blond. 
Her hair is painted a bright yellow stippled with fine red horizontal 
lines, 23 and her skin is white. This is the earliest known evidence of blond- 
ism in the world. Later Egyptian reliefs, however, frequently represented 
Libyans as blond, 24 and in early Egyptian times, the territory of the 
Libyans extended to the Delta itself. The Egyptian representation of 
foreigners is quite accurate; besides the Libyans, who have Nordic features 
as well as coloring, Asiatics, with prominent noses and curly hair, sea 
peoples from the Mediterranean, with lighter skins and a more pronounced 
facial relief than the Egyptians, are also shown, as well as negroes. 

The blondism of Hetep-Heres II apparently belonged to the Delta and 
to the connections outside to east or west, rather than to Egypt proper, for 
it never recurred as an important or characteristic Egyptian trait. The 
Mediterranean pigmentation of the Egyptians has probably not greatly 
changed during the last five thousand years. 


In view of the importance of North Africa as one of the two main cor- 
ridors of Neolithic diffusion into Europe, it is extremely disappointing that 
in it very few human remains of this cultural period have been found. A 
handful of skulls from Redeyef and Tebessa, near the border between 
Algeria and Tunisia, are the only surely Neolithic ones that have been 
described. 26 These are all of the small-statured, thin-boned, small-headed, 
dolicho- to low mesocephalic variety of Mediterranean already seen at 
Muge; smaller, on the whole, than most of tile early Egyptians, and 

2 ' Reisner, G. A., BBMF, vol. 25, #151, October, 1927, pp. 64-79. 

84 Bates, O., The Eastern Libyans. 

25 Bertholon and Chantre, Rec her ekes anthropologiques dans la Berberie Oriental*, pp. 237- 


shorter-headed than the small Badarians. They cannot be derived directly 
from Egypt proper, nor from any known population of the Delta, if the few 
Merimdian skeletons already mentioned may be considered typical of that 
region. This small and geographically limited group is a local form of 
Mediterranean of the same variety which, at a presumably earlier date, had 
crossed the Straits into the Iberian Peninsula. 

Other remains, found in caves in eastern Algeria, 26 are likewise small 
in absolute body size, having a mean stature of approximately 160 cm., 
but resemble the type of T6viec rather than that of Muge. They may be 
attenuated Afalou survivors, but cannot with certainty be ascribed to the 
Neolithic. Many, if not all, may be Mesolithic in date. 27 

The megalithic cultural complex, borne through the Mediterranean by 
sea in the Late Neolithic, and spreading northward past Gibraltar to the 
British Isles, France, and Scandinavia, reached the North African shores. 
But in this minor theater of megalithic activities the stone monuments, 
which do not occur east of Tunisia, may have been first erected in post- 
Neolithic times, since most of them contain objects of bronze, or even of 
iron. They were, in fact, occasionally used as burial vaults through Roman 
times, and right up until the arrival of the Moslems. Under these circum- 
stances we cannot expect to find a purely megalithic race in the Tunisian 
and Algerian dolmens M and, to a certain extent, the material lives up to 
expectations. Although the cranial indices, in some thirty specimens, 
ranges from 67 to 84, the majority of the skulls are dolichocephalic, and 
some of them are extremely long, while most of them are leptorrhine, 
unlike the broader-nosed ordinary Mediterranean crania of the Neolithic. 
Furthermore, the stature of the dolmen people is tall, with a male mean of 
about 168 cm. 29 Unless these are the skeletons of Hamites or Arabs, we 
may infer that the megalith builders were not the small Mediterraneans 
proper of Mesolithic tradition, but a new ethnic element which we shall 
be able to study more profitably when we find it in greater numbers farther 
to the north. 


It is not easy, from a distance, to collect and review the evidence for the 
Neolithic population of the Iberian Peninsula. I have been able to assem- 
ble data on some fifty crania from Spain, and nine from Portugal, which 
seem, with reasonable certainty, to be of Neolithic age. 80 

26 Ibid., pp. 240-242. 

27 Boule, M., Verneau, R., Vallois, H., AIPH, Mem. 13, p. 190. 

28 There are very few in Morocco, and nothing is known of their skeletal contents. 

29 Bertholon and Chantre, op. cit. t pp. 243-249. 

80 Scheldt, W., in his Die Rassen der jungeren Stein&U in N. W. Europa, pp. 87-92, ac- 
cepted but 38, besides the 68 early Bronze Age crania from el Argar. Gzortkower, S., 
the author of another compilation (PAn, vol. 8, 1934, pp. 45-52), used 118 from Spain, 


The Portuguese specimens, all from the Tagus Valley, can all be classi- 
fied as Mediterranean. They include, however, not only the small Muge 
type, but others with larger skulls and taller stature, as high as 168 cm. 
in the case of one male. 31 

The Spanish material is best represented by two series, the first from the 
cave of La Solana at Angostura, Segovia 82 (see Appendix I, col. 9), and 
the second from the cave of Ticuso at Sepulveda, in the same province. 33 
Both of these series were originally called Magdalenian, but the presence 
of pottery and polished celts in the Solana cave, and of trephination at 
Ticuso, leave little doubt that both are really Neolithic. 

The Solana series, which includes ten males and four females, repre- 
sents a relatively large Mediterranean type, which may be nearly duplicated 
in the Egyptian series from the royal tombs at Abydos 34 arid would also fit 
metrically into a Mesopotamian Eurafrican type group. Morphologically, 
the crania are relatively heavy, with moderately large supraorbitals. 

The second series, that of Ticuso (see Appendix I, col. 10), includes 
fourteen male and seven female crania. These are somewhat smaller and 
more delicately formed than the Solana series, and resemble metrically the 
Naqada predynastic skulls from Upper Egypt. Smaller series and single 
skulls from other parts of Spain usually fall into this same category. 

The human remains which represent the Neolithic period in Por- 
tugal and Spain, therefore, incomplete as they are, corroborate the 
evidence of archaeology. The Iberian Peninsula was a corridor of 
movements into western Europe from North Africa, and two types, at 
least, made use of this passageway a small variety of Mediterra- 
nean, somewhat larger than the Mesolithic people of Muge, but bas- 
ically the same, and identical with the people who moved into the upper 
valley of the Nile in predynastic times; and a somewhat larger, heavier 
sub-division of the same race, similar to Neolithic man in western 
Asia, and perhaps to the early farmers of the Egyptian Delta. To 
what extent these two types included local Mesolithic survivors it is im- 
possible to tell. 

which probably include el Argar. When these are subtracted his list attains exactly the 
same size as mine. 

81 Barros e Cunha, J.-G. D., ACIA, 3me Session, Amsterdam, 1927, pp. 358-360. 

Herve", G., REAP, vol. 9, 1899, pp. 265-280. 

Mendes-Correa, A., BAC, vol. 3, 1925, pp. 117-146. 

Herv6 states (p. 274) that the series includes a few brachycephals, but the published 
data do not support this. 

Barras de Aragon, F. de las, AMSE, vol. 12, 1933, Cuad. 1, pp. 90-123; Verneau, 
RDAP, 1886, ser. 3, vol. 1, pp. 10-24. 

^Hoyos Sainz, L., CRCA, 14me Sess., Geneva, 1912, vol. 2, pp. 399-408; Barras 
de Aragon, ibid. 

** Morant, op. cit., 1925. 



North of the Pyrenees, the Neolithic population of Europe was im- 
mediately derived not only from Africa, but also from the east. In order 
to understand the racial complications of trans-Pyrenean Europe in the 
Neolithic, we must converge from a different quarter. The eastern source 
areas, and their possible routes into Europe, may be divided into three: 
(a) Crete and the Aegean Islands, thence by sea to Greece, and to Italy, 
and from Greece, northward by land into Macedonia, (b) From Anatolia 
over the Bosporus into the Balkans, and thence up the Vardar and down 
the Morava into the Danube above the Iron Gates, (c) Around the north- 
ern shore of the Black Sea, and perhaps of the Caspian Sea as well, then 
the steppes of southern Russia into the plains which reach through Poland 
to Germany, and into the Danube Valley. 

(a) Our knowledge of the physical type in Greece during the Neolithic 
is confined to one small, narrow, female skull of Mediterranean type, from 
Arcadia, 35 which, as we shall soon see, is perfectly consistent with the 
racial picture farther north, although it is not very likely 36 that racial 
movements passed northward from this quarter at that time. Crete, whose 
civilization was rooted in the Neolithic, is unknown racially until the 
Bronze Age. 

The Neolithic inhabitants of Italy probably came from the east in large 
measure by sea, although some may have entered from other directions, as 
from North Africa by way of Malta and Sicily, around the Tyrrhenian 
Sea from Catalonia, and down over the Alps from the north. 

It is also very likely that Mesolithic types, containing an earlier Palaeo- 
lithic increment, survived in Italy into the Neolithic, for, until the arrival 
Df metal, Italy and its islands formed an area of relative isolation from the 
main racial and cultural currents which affected Europe as a whole. 

Although Aeneolithic or Copper Age skeletons from Italy are abundant, 
those dating from Neolithic time are rare. 37 All that have been found 38 
(51) are long-headed, and of Mediterranean type. Three skulls from the 
Ligurian cave of Arena Candide which are very large and of great length, 
may represent, at least in part, an Upper Palaeolithic survival of Early 
Aurignacian type, or an invasion of the tall Mediterranean type usually 
identified with the megalith-builders. It will be more profitable, however, 
to defer the study of racial types in early Italy and her islands until our 
discussion of the Copper and Bronze Age population, when we shall have 
something more definite and extensive with which to work. 

86 Fiirst, Carl M., LUA, NF. Avd 2, Bd. 28, #13, 1932. 

M Fewkes, V. J., Goldman, H., Ehrich, R. W., BASF, #9, 1933, p. 18. 

87 Sergi, G., Europa, pp. 270-289. 

88 With the exception of one microcephalic skull, op. cit., p. 279. 


(b) The second eastern source area from which Neolithic invaders 
may have entered Europe is that of the Anatolian plateau to what ex- 
tent the Danubian peasants were derived from these highlands is a matter 
of dispute among archaeologists which we shall do well not to enter. At 
any rate, no Neolithic skeletal remains have yet been found there, and the 
metal period sites which have been studied are later than those in Mesopo- 
tamia. Farther east, at a site called Zizernakaberd in Armenia, the brain 
case of a tall man (172 cm.) with apparently Upper Palaeolithic affinities, 
resembling Murzak Koba, may have been buried in the earliest Neo- 
lithic time. 89 This one specimen from Armenia is small evidence, and we 
still do not know what kind of people lived in Anatolia at the time when 
the first farmers pioneered up the valley of the Danube. 

(c) The third eastern source area, and perhaps the most important of 
the three in the total peopling of Europe in the Neolithic and later, is the 
grassy plain extending from Poland across Ukraine and Bessarabia, north 
of the Black Sea and Caucasus, across to the Caspian, and beyond into 
Turkestan. Here the evidence of Neolithic man is considerably better than 
in the other two. 

On the eastern side of the Caspian, near the modern border between 
Russian territory and Iran, are the three famous Kurgans, or mounds, of 
Anau. The earliest cultural horizon found in this site, Anau I of the north 
mound, probably dates from 3500 to 3000 B.C., on a conservative estimate. 
This level, which is largely but not purely Neolithic, contained a number 
of human skeletons, 40 most of which were those of children. 

All of the children were dolichocephalic, and apparently of Mediter- 
ranean type. One adult female, found with them, was the same. She was 
mesocephalic, with a cranial index of 76, and her skull shows a minimum 
of bony relief. The forehead projects forward, the glabella is almost absent, 
the nasal root high, and the nasal profile apparently straight; the orbits 
are mesoconch, and the facial bones delicate. 

Another adult, in this case a male, is represented by a mandible and 
certain facial bones below nasion. Again a Mediterranean type is indi- 
cated, orthognathous, with a strong lower jaw, and a small nose which was 
moderately leptorrhine. This specimen, the female, and the children, al- 
though hardly a series, are sufficient to show us that this southwestern 
corner of Turkestan was inhabited by agricultural, animal-breeding, 

Vishnevsky, B. N., MAGW, vol. 64, 1934, pp. 102-111. 

40 Moliison, T., "Some Human Rernains Found in the North Kurgan, Anau," in 
Pumpelly, R., Explorations in Turkestan, vol. 2, pp. 449-463. 

Sergi, G., "Description of Some Skulls from the North Kurgan, Anau," ibid., pp. 445- 
448; ASRA, #13, 1917, pp. 305-321. 

Warner, Langdon, "Report on Skeletons Excavated at Anau," in Pumpelly, R., 
op. cit., p. 484. 


pottery-making people of general Mediterranean type in the second half of 
the fourth millennium B.C., as early as the predynastic period in Mesopo- 

Long bones from the following level in the North Kurgan show varia- 
tions in stature with two males at 170 and 161 cm., respectively, and a 
female at 149 cm. 

A post-Neolithic skull from the South Kurgan, probably of the third 
millennium, is, like the others, dolichocephalic. It has a low, sharply 
curved forehead, no browridges, small zygomatic arches, and apparently 
considerable prognathism; 41 but an exact racial diagnosis of it cannot be 

Returning to the Neolithic material, we may be sure that it all belongs 
to some branch of the Mediterranean race, but, with the present evidence, 
which does not contain a single complete adult male specimen, we cannot 
hope to distinguish the skeletal sub-variety. 

In the grasslands of European Russia, south of the forest belt, a racial 
continuity with Anau extends westward into the Ukraine. One of the 
earliest sites which show this connection is located at Mariupol near the 
mouth of the Kalmins River on the shore of the Sea of Azov. 42 Here, an 
unstated number of skeletons, lying in rows and covered with red ochre, 
was found in association with apparently Early Neolithic implements, and 
a quantity of bone, shell, and tusk objects. Although the typology of the 
artefacts is early, we do not know the date, but the absence of pottery 
would presumably argue against a late assignment. 

No measurements of these skeletons have been published, but the de- 
scription is sufficient to show that a Mediterranean type, perhaps similar 
to that found at Anau, is probably involved. The stature was "slightly 
above the medium height of today," 43 which would place it in the upper 
160's; the bones of the extremities are elongated, the hands narrow and 
long. The skulls are small, and in all cases dolicho- or mesocephalic. 

Neolithic crania from southwestern Russia and the adjacent segment 
of Poland are not numerous, but are clearly differentiated racially. 44 They 
belong to two types; a high-vaulted, moderately broad-nosed dolicho- to 
mesocephal, associated with short stature, 160 cm. or less, in the males. 
This type, which carries the Anau form to the west, is the most numerous, 

41 From a poorly oriented photograph given Sergi by Pumpelly and published by the 
former, without measurements. Sergi, G., ASRA, vol. 13, 1907, pp. 305-321. 

42 Makarenko, N., ESA, vol. 9, 1934, pp. 135-153. 
/*/., p. 140. 

44 Bogdanov, A. P., AAM, vol. 3, 1879, part 1, p. 305. 

Gzarnowski, S. J., Swiatowit, vol. 3, 1901, pp. 75-84. 

Levit'kyj, I., AntrM, vol. 2, 1928, pp. 192-222; ZVAK, vol. 1, 1930, pp. 159-178. 

Sailer, K., AAnz, vol. 2, 1925, pp. 26-46. 

Zabrowski, S,, BMSA, ser. 5, vol. 2, 1901, pp. 640-666. 


and is centered in the Volhyn district of the Ukraine. With it, in the Late 
Neolithic Fatjanovo culture, are associated a few brachycephals which, 
except for head form, differ little from the rest. This "Danubian" type is 
not basically different from some of the Lower Egyptian and Delta groups. 

The second type, commonest in Late Neolithic cemeteries of the Kiev 
government, is of the tall (stature = 171-1 72 cm.), hyperdolichocephalic 
variety, usually leptorrhine and high-vaulted, which we have called 
" Corded." Crania of this variety are actually few in number, and prob- 
ably Late Neolithic in date. Metrically, they resemble the earliest Sume- 
rian skulls at el Ubaid. 

Sergi, on a visit to Moscow some thirty years ago, measured over 
seventy male " Kurgan" crania from southern Russia, dating from all 
periods from the Neolithic to the pre-Christian Iron Age. These, selected 
as "Mediterraneans," 45 conform to the two types mentioned above. The 
main group, the smaller variety, fits our "Danubian" type, the larger, 
the "Corded." In general, the metrical deviation of the total group from 
Mesopotamian figures is not great. 

The result of this south Russian inquiry leads to several cumulative if 
tentative conclusions: 

(1) During the Neolithic, all known avenues of approach to Europe, 
from Gibraltar to the southern limit of the Russian forest, show only 
variants of Mediterranean or Galley Hill man. The Neolithic culture 
with its food-producing economy, and the Mediterranean race, are, as 
Sergi said, inseparably linked. 

(2) The special "Mediterranean" form, which had apparently brought 
agriculture to the countries north of the Iranian plateau and Black Sea, 
was not unlike others found in more southerly regions in which Old 
World agriculture is supposed to have originated. 

(3) The tall, hyperdolichocephalic high-vaulted variant of the basic 
Galley Hill stock, elsewhere to appear as the Corded people, was present, 
at least by the Late Neolithic, in southern Russia. 


One of the most striking events of the Neolithic period in Europe was 
the gradual migration of farmers up the Danube Valley into central 
Europe. These new settlers stayed fairly close to the banks of the river 
and its tributaries, farming on patches of loess where the land would not 
need to be cleared by the axe. Southern Hungary, Moravia, Bohemia, 
and Silesia were areas which they found especially favorable, and in which 
they settled in greatest numbers. As they moved to the west, they finally 

45 Sergi, G., Europa, pp. 309-316. In Sergi's own words, Eurafrican. This term has 
since taken on a narrower meaning in the hands of Mesopotamian archaeologists. 


reached southern Bavaria, Baden, and the north of France, especially 
the Paris basin. From southern Germany onward, they encountered the 
descendants of the Neolithic people who had entered by way of Gibraltar. 

The river valleys which the Danubians occupied must have been rela- 
tively free of people; Mesolithic remains in the eastern and middle Danube 
Valley are very scarce, if not entirely absent. 46 We may therefore expect 
the remains of the Danubian immigrants to exhibit, without particular 
alteration, the physical characteristics of the population or populations 
from which they originated. 

Danubian chronology is based on pottery types, particularly on tech- 
niques of decoration; the earliest Danubian, Period I, is typified by in- 
cised pottery with banded decoration, while the second and third periods 
mark the common use of painted pottery. The agriculture of the Danu- 
bians was a hoe-culture, for the characteristic tool is a hoe blade of flint, 
called a "shoe-last celt." Their domestic animals included the ox, sheep, 
and pig. 

It is one of the problems which face the archaeologist in the future to 
discover the point of origin of Danubian pottery. Incised black ware, of 
the banded variety, undoubtedly came from somewhere to the east; from 
the country north of the Black Sea, or from Anatolia, whence it may have 
been influenced by the same source which produced the Merimdian of 
the Egyptian Delta. In this case, the two movements, the Danubian and 
that which passed over Gibraltar, may have come from a single original 
source in western Asia, and have moved into Europe from two different 
directions, converging in Switzerland, southern Germany, and France. 

The painted pottery, on the other hand, shows definite Asiatic simi- 
larities; there was painted pottery in Iraq in the earliest known cultures; 
Anatolia contains some varieties of it; the Iranian plateau is said to be 
full of it; there is painted pottery at Anau in Turkestan; and painted 
pottery penetrated early into Kansu in China. Despite these occurrences, 
we do not yet know by which route or routes it entered Europe from the 
east. It may have come across the Bosporus, around the Black Sea, or 
from both quarters. Again, it may have travelled, farther east, either 
north or south of the Caspian. 

The physical evidence at hand will hardly settle the problem of Danu- 
bian origins, although it will, in a fragmentary manner, dispel a number of 
unfounded hypotheses. In the material used in the present survey, seven- 
teen male crania associated with banded pottery, 47 and seven associated 

"Fewkes, V. J., Goldman, H., Ehrich, R. W., BASF, #9, 1933, pp. 17-32. Also, 
personal communication of Dr. V. J. Fewkes. 
* 7 Bayer, J., MAGW, vol. 51, 1921, pp. 46-47. 

Lebzelter, V., MAGW, vol. 66, 1936, pp. 14-15; ibid., "Sitzungberichte," p. 16. 
Reche, O., AFA, vol. 35, 1908, pp. 232-237. 


with painted, 48 are all that can without doubt be attributed to the Danu- 
bian Neolithic^ These may be supplemented by a smaller female series. 

The two series, Banded and Painted, are so close to each other anthropo- 
metrically that they may readily be pooled (see Appendix I, col. 11). Their 
type is a familiar one a small Mediterranean, with cephalic indices 
ranging from 68 to 81, and a mean of 73.6. The mean cranial length is 
185.5 mm., but individually they go as high as 196 mm. The vault height, 
139 mm. is elevated in comparison to the other dimensions. The faces 
are short (116 mm.), and moderately narrow (130 mm.); both foreheads 
and jaws (minimum frontal 96 mm., bigonial 94 mm.) are also of moderate 
breadth. The orbits are low, with an orbital index mean of 80, the noses 
chamaerrhine, with a nasal index mean of 55. The highest orbitted skull 
has an orbital index of 91, the most leptorrhine a nasal index of 45. 

Although this Danubian group is reasonably homogeneous, even with 
the small numbers available it is seen to include more than one type in 
the strictest sense. For example, the stature is low; Reche found a mean 
of 153 cm. for eight Banded male skeletons from Jordansmuhl, and in this 
small series four mesocephalic crania are associated with higher statures 
than are the purely dolichocephalic ones. Some of the skulls with higher 
orbits and longer vaults differ again from the majority. On the whole, 
however, the group is definitely dolicho- to mesocephalic, and definitely 
Mediterranean. As far as the criteria studied may be invoked, this series 
is very similar to Sergi's Kurgan group from southern Russia, and may be 
considered to contain the same racial elements, although the Russian mate- 
rial as a whole is less homogeneous. 

If we carry the comparison further, we find, again, strong resemblances 
in the Spanish Neolithic, and with all of the smaller Mediterranean 
groups. The Danubians undoubtedly represent another branch of the 
same racial group which entered Europe from North Africa through the 
southwestern avenue. Where they came from immediately before their 
arrival in Europe, however, it is impossible at the moment to tell. The 
Russian evidence, including that from Mariupol and Anau, leans heavily 
in favor of a trans-Euxine origin, but at the same time they might have 
come from Anatolia, from which we have as yet no Neolithic skeletal 
evidence. It is again possible that related elements from more than one 
geographical source made up the Danubian migrations. 

We do not know what language the Danubians spoke, nor what was 

H, A., ACAP, 1931, pp. 114-115. 
Lebzelter, V., WPZ, vol. 15, 1928, pp. 35-41. 
Nestor, I., BRGK, #119, 1933, p. 37. 

Schiirer von Waldheim, Hella, MAGW, vols. 48-49, 1919, pp. 247-263. 
Virchow, R., ZFE, vol. 22, 1890, p. 97. 
Zimmerman, G., AJKS, vol. 10, 1935, pp. 227-236. 


the coloring of their skin, hair, and eyes. But we may surmise from the 
small evidence which has been assembled that the successive waves repre- 
sented did not come from racially different parent groups. 

Although we cannot, from this evidence, state what racial elements 
were lacking in the Danubian countries during the Neolithic, we know 
that the culture bearers from the east belonged to, or included mem- 
bers of, the wider Mediterranean stock, which seems everywhere to be 
associated with the earliest food production; and the most important ele- 
ment seems to have been a small, light boned, rather infantile Mediter- 


The latter part of the Neolithic period in most of north central Europe 
is marked by the appearance of an enigmatical group of people, who 
decorated their pottery, while still wet, with cord impressions, and who 
also placed in their graves perforated stone battle-axes suspiciously like 
those of the Fatjenovo culture in southern Russia, and others in the Cau- 
casus. These axes, again, have copper parallels in Sumeria. The limits 
of the country overrun by the Corded people are the Vosges on the west, 
the Urals on the east, the Baltic on the north, and the Dinaric Alps on the 
south. 49 Although these invaders were partly agricultural, their graves 
contain weapons rather than hoes, and, in a few cases, bones of horses, 
probably of a domestic variety. 

Their r61e in the economic and political picture of Neolithic Europe 
remains still in doubt. Although they were equipped for warfare, they 
did not fight for the love of battle alone. The location of their burying 
grounds near the sources of natural wealth, such as amber, salt, and later 
of tin, shows that they were interested in easily traded commodities of 
small bulk but high value. They may have been Neolithic racketeers 
extorting their share from the drones, or overlords among peasants, or 
merely industrious and well-armed peddjers. Whatever their calling, 
whether peaceful or otherwise, they were destined to influence the later 
cultures of Europe in considerable degree. 

The most typical aggregation of Corded skulls comes from Silesia and 
Bohemia, whence a series of twenty-nine males may be assembled. 50 (See 
Appendix I, col. 12.) These belong to a very definite, very distinct physical 
type. The length of the vault is great, well over 190 mm. in most instances; 
its breadth is slight, yielding the low mean cranial index of 71; and the 
height is great, considerably exceeding the breadth. Combined with this 

49 Childe, V. G., The Danube in Prehistory, pp, 145-160. 
60 Rechc, O., AFA, vol. 35, 1908, pp. 232-237. 
Stock^, A., AnthPr, vol. 7, 1929, pp. 65-78. 


exaggeratedly long, narrow, and high vault form is usually found a high, 
relatively steep forehead; stronger browridges and muscular markings than 
are usual with the Mediterranean types familiar to us in Egypt, Spain, and 
the Danube; while the face form includes compressed zygomata, low or- 
bits, and a leptorrhine nose. The face heights are probably great, and the 
mandible is deep and strongly marked, although usually narrow. Unfor- 
tunately, in this series, these facial descriptions are much less certain than 
those of the vault, for few of the crania retain their facial segments. The 
long bones are heavier and more rugged than those of the smaller Mediter- 
ranean varieties, but the stature, ranging between 157 and 170 cm. in 
ten male examples, reaches the unimpressive mean of 164 cm. In other 
Corded series, as we shall see later, it is almost always tall. 

The Corded crania are larger than any from Egypt, and are metrically 
very similar to the Elmenteita skulls from East Africa the two groups 
could be combined without loss of homogeneity. In Mesopotamia, they 
may be favorably compared with the three early dynastic skulls from Ur, 
although they are higher vaulted than the other early groups. 

There has been much discussion over the origin of the Corded people, 
and many cradle-areas have been proposed. Childe, despite several ob- 
jections which he himself raises, prefers to derive them from southern 
Russia, where the typical cultural elements of the Corded people are 
found mixed with other factors. The so-called boat-axe, the typical 
battle-axe form which they used, has relatives all the way to the Caucasus 
and beyond. And the horse, their use of which in the domestic form is 
not fully confirmed, since the grave examples might conceivably have 
been wild ones, was first tamed in Asia or in southern Russia. 

On the basis of the physical evidence as well, it is likely that the Corded 
people came from somewhere north or east of the Black Sea. The fully 
Neolithic crania from southern Russia which we have just studied include 
such a type, also seen in the midst of Sergi's Kurgan aggregation. Until 
better evidence is produced from elsewhere, we are entitled to consider 
southern Russia the most likely way station from which the Corded people 
moved westward. 

There is one cautionary remark which must be made here, and that is: 
there is so far no justifiable reason for assuming that the Corded people were 
Nordics. Their cranial type, as we know it, does approach one or more of 
the forms which we know, in later times, to have been associated with 
blondism; but it also approaches those of the Iranian plateau and of Ur, 
which were probably brunet. Let us withhold judgment, therefore, upon 
Corded soft parts and pigmentation, and view these remains in the more 
scientific but less lively light of a skeletal type. 

This Corded skeletal type is familiar also in Poland, where it is found 


in the graves of its associated culture; but that country also contains the 
more usual Danubian type, associated with a Neolithic agricultural 
economy, and a certain number of brachycephalic and other crania, 
which have northern affiliations, and which will therefore be dealt with 
later. 51 

In southern and western Germany remains of the Corded people are 
again found, and in comparative abundance. In Saxony and Thuringia 
they flourished especially, and apparently were more stable here than 
farther east. Out of ten crania which belong to the Saxo-Thuringian 
Corded culture, 62 four of the seven which can be measured are meso- 
cephalic, and only three dolichocephalic. In the eastern Corded group, 
the highest index was 75. The three dolichocephals seem to have be- 
longed to the usual type. 

The statures of two of them were both 1 68 cm. The rest of the crania, 
as far as one can tell, are normal Neolithic Mediterranean examples, 
which might have had either a Danubian or a North African derivation, 
or both. The Corded people in the west and south of Germany had 
settled down, and had combined with Neolithic farmers. 

Before we leave this section, let us move still farther west to Baden, to 
the Early Neolithic cemetery of Altenburg. 53 Here, in the center of one of 
the most brachycephalic regions of Europe today, were buried four male 
skeletons, the crania of which ranged from 65 to 71 in cranial indices, and 
two female skulls of 77. The long bones are small, the statures short; the 
skulls are delicate in appearance and purely Mediterranean but remark- 
able for the narrow vault form of the males. Six other Neolithic male 
crania, from Worms, are similar. 54 This evidence, while not complete, 
at least shows that the Corded people, in southern and southwestern Ger- 
many, were preceded by an agricultural population of the smaller Medi- 
terranean variety, upon which they superimposed themselves. 


The next move in this geographical game is back to the extreme west 
again, and to Britain. The Early Neolithic culture of the British Isles was 
a peripheral echo of the movements which influenced the rest of western 

51 Lcncewicz, Stanislaw, Swiatowit, vol. 10, 1912, pp. 53-64. 

Rosinski, B., WArc, vol. 9, 1924-25, pp. 29-50; ACIA, 2me Session, Prague, 1929, 
pp. 164-174. 

Westlawawa, Eleanora, PAn, vol. 9, 1935, pp. 80-84, French r6sum6, pp. 142-143. 

62 Gotze, W., JVST, vol. 24, 1936, pp. 91-100. 

Heberer, G., JVST, vol. 24, 1936, pp. 82-90. 

Strauch, K., MannusZ., vol. 7, 1915, pp. 249-262. 

68 Miihlmann, Wm. E., ZFMA, vol. 28, 1930, pp. 244-255. 

" Virchow, R., ZFE, vol. 29, 1897, p. 464. 


Europe. The so-called Windmill Hill culture, closely allied to the Michels- 
burg expression ill southern Germany, may have been originally of either 
North African or Danubian inspiration, or a blend of both. Childe, seeing 
Merimdian similarities in the pottery, suggests but does not insist on the 
former. At any rate, we have no valid evidence in Britain itself to indicate 
the physical type of the people who brought it. 55 

The bulk of the Neolithic population of the British Isles seems to have 
come by sea, 66 with the Megalithic invasions which also passed on to 
Denmark and southern Sweden. In many parts of Scotland and in Ire- 
land, the Megalithic people may well have been the first bringers of the 
Neolithic economy. In England, it was their custom to make primary 
interments under long barrows of earth, unchambered in Yorkshire and 
Derbyshire, chambered in the counties farther south. 

The cranial remains of Long Barrow men, as the occupants of these 
monuments are called, are abundant. 67 (See Appendix I, col. 13.) Al- 
though over 160 skulls represent this group, the geographical distribution 
is far from even. Wiltshire, Staffordshire, and Gloucestershire account for 
120; fourteen only are from Scotland, and one from Ireland. The remain- 
ing thirty come from a few counties of England. Wales is unrepresented as 
is most of Scotland; the few crania found in the latter country were all 
buried close to the sea. The Long Barrow people, who had come by water, 
selected open, unforested country to live in. A large .part of the land area 
in the British Isles was, therefore, either uninhabited or open to the wan- 
derings of earlier human occupants. 

The Long Barrow population formed a distinct, homogeneous type; 
one different from any which, to our knowledge, had previously inhab- 
ited the British Isles since the days of Galley Hill; and one which cannot 
be duplicated, except as an element in a mixed population, anywhere on 
the western European continent. One is, therefore, led to conclude that 
the Megalithic cult was not merely a complex of burial rites which dif- 

56 The so-called river-bed skulls, dredged from the bottom of the Thames, are those 
of low-vaulted Mediterraneans. These may include some examples from the Early Neo- 
lithic, but the evidence is inconclusive. (Garson, J. G., JRAI, vol. 20, 1890, pp. 20-25.) 
Three skulls from stone cists at La Motte, Jersey are similar. (Marett, R. R., Archae- 
ologia, vol. 63, 1911-12, pp. 203-230. Keith, Sir A., Antiquity of Man, vol. 1, pp. 

66 Childe, who read Chapters II to VII in manuscript before revision, comments at 
this point: "I find it hard to believe that the bulk of the British population came by sea. 
The Windmill Hill culture is predominant in the megalithic tombs, but arose earlier." 
While Childe is undoubtedly correct as to the importance of the Windmill Hill people 
culturally, there is little evidence of them in a physical sensf This apparent contradic- 
tion cannot be explained on the basis of present data. The fact that small Mediter- 
raneans do appear in the living British population (see Chapter X) indicates that 
Childe's observation may be well founded. 

67 Morant, G. M., Biometrika, vol. 18, 1926, pp. 56-98. 


fused without visible carriers; and also that the bearers of this complex 
avoided mixture by coming by sea. 

In stature and bodily build, the Megalithic people belong to a large 
variety of Mediterranean. The stature for a large number of males 58 
from England ranges about a mean of 167 or 168 cm.; which is not con- 
traverted by the meager evidence from Scotland and Ireland. Four male 
skeletons from a single burial in Kent 69 may represent, more nearly than 
most, the Windmill Hill group; they are somewhat shorter than the rest. 

The Long Barrow skulls are large for a Mediterranean sub-race, but 
not as large as those of the Upper Palaeolithic peoples. They are par- 
ticularly long, moderately narrow, and of medium height. Unlike that of 
the Corded skulls, the height is less than the breadth. In most instances, 
the occiput projects far to the rear; the parietals are parallel; the forehead 
is moderately sloping, and, in contrast to the restricted skull width, very 
straight and broad. 

The face is of medium length and of moderate width; the orbits are of 
medium dimensions, and in many instances slope downward an4 outward, 
as if the confines of the face were too narrow for them. The nasion depres- 
sion is of medium depth, under browridges of medium development; and 
the straight-profiled nose is leptorrhine. In its totality, the I^ong Barrow 
type is both extreme and striking. 

In looking for related populations of equal age, we may eliminate at 
once the smaller, less dolichocephalic branches of the Mediterranean race 
proper, including the Danubian. A few individual crania in Neolithic 
Spain and Italy would qualify, but none of the series from these coun- 
tries. The standard Egyptian crania, as groups, are all too small, as is 
the single lady from Greece. In one particular feature, the nasal index, the 
Long Barrow people resemble the Egyptians more than most of the more 
northerly Mediterraneans, for the Long Barrow crania are leptorrhine. 

In their extreme dolichocephaly, the Long Barrow skulls resemble the 
Corded group, but the comparison does not hold for all features the 
Long Barrow skulls are slightly longer', considerably broader, and much 
wider of forehead, than the Corded specimens, and, of course, the vault 
of the Long Barrow skulls is much lower. 60 As far as one can tell, the 

88 Calculated by the Pearson formulae on femora from several series, including some 
eighty-six individuals from England, of which many may be duplicates; three from 
Scotland, and one from Ireland. Sources: Crania, Britamuca; Thurman, J.; Garson, 
J. G.; Mortimer, J. R.; Keith and Bennett; Edwards, A. J. H., and Low, A.; Laing, S., 
and Huxley, T. H.; and Bryce. 

69 Keith, Sir A., and Bennett, JRAI, vol. 43, 1910, pp. 86-100. 

60 In this I am relying on Morant's mean of 135.5 tnm. for 25 male crania. Schuster 
(1905) gives 137.8 mm. for 12; Garrison, 135.0 mm. for four from Howe Hill Barrow, 
Yorkshire. On the other hand, 45 male crania of Thurman (1867) when seriated 
143 mm., 59 from the Crania Britannica and Thurman = 142.1 mm. 


orbits in the two series are much the same, while in regard to the faces, 
there is not enough evidence in the Corded group for a valid comparison. 

A true and valid similarity, however, may be found between the Eng- 
lish Long Barrow series and the early skulls from al 'Ubaid in Sumeria, 
which, whether belonging to the fourth or third millennium B.C., are in 
either case older than their British counterparts. The only difference, 
which prevents identity, is that the Mesopotamian faces and noses are 
somewhat longer. 

The current idea that the Long Barrow people were directly derived 
from the Upper Palaeolithic inhabitants of Britain is clearly erroneous. 
The Long Barrow skulls are definitely smaller, shorter, and narrower than 
those of the Upper Palaeolithic group, but of equal or greater height; 
they have the same forehead breadth, the same upper face height, but a 
smaller jaw, a much narrower face, and narrower orbits. There is prob- 
ably a genetic linkage, over a long period of time, between the Long Bar- 
row or Megalithic type and an early Galley Hill or Combe Capelle vari- 
ety of European man, but the continuity could not, for historical reasons, 
have taken place in England. 

The few crania from the Scottish seashores belong to the standard Long 
Barrow type, and the same may be said of the one surely Neolithic spec- 
imen from Ireland the male vault from Stoneyisland, Portumna, County 
Galway. 61 The male skull from Ringabella, County Cork, 62 which is 
perhaps also Neolithic, is likewise of Megalithic race, while the disputed 
Kilgreany specimen, whatever its age, is, although low vaulted, also basi- 
cally of a Galley Hill Mediterranean type. 63 However, the large mandible 
of the latter, and its low vault, make it atypical, so that it, like two skulls 
from Phoenix Park, Dublin, 64 which may be Neolithic or Early Bronze 
Age, is not wholly characteristic of the Long Barrow race, and may derive 
its peculiarities from either a Mesolithic or an Early Bronze Age source. 
We must repeat, in view of these aberrances, that the only surely Neolithic 
skull in Ireland is of Long Barrow race. 

The Megalithic Long Barrow people must have come by sea, and they 
probably came from somewhere in the Mediterranean. They did not 

61 Martin, G. P., JSAI, vol. 64, June, 1934, pp. 87-89. 

Movius, H. L., Jr., op. cit. 9 vol. 65, Dec., 1935, p. 282. For dating by palaeobotany, 
see Shea, S., JGAS, vol. 15, 1931, pp. 73 ff. 

White, Miss J. M., INF, vol. 3, 1934, pp. 270-274. 

62 Martin, G. P., in 6 Riordain, S. P., JSAI, vol. 64, June, 1934, pp. 86-87. 
68 Fawcett, E., PBSS for 1928, vol. 3, #3, pp. 126-133. 

Martin, G. P., as above. 
Movius, H. L., Jr., as above. 
Tratman, E. K., ibid., pp. 134-136. 

* 4 Haddon, A. G., PRIA, vols. 3, 4, 1896-98, pp. 570-585. Also, Crania Britannica, 
skulls 22 A and B. 


find the British Isles uninhabited, and their homogeneity, in a few re- 
stricted localities, cannot mean that they caused the extinction of earlier 
peoples. Nor did they, when still later invasions of another physical com- 
plex reached the British Isles, become extinct. 66 The mountains of Wales, 
the hills of Cornwall and Devon, and almost the whole of Ireland, remain 
a blank in our early skeletal map of the British Isles. 


By this time we have studied all of the approaches by which Neolithic 
food-producers invaded Europe, and have seen that in all known cases 
these immigrants belonged to some branch of the Galley Hill stock or 
wider Mediterranean race. We now come to the portions of Europe in 
which the Mesolithic cultural tradition had a strong survival as a blend 
into the Neolithic economy, or as an absolute continuation. These por- 
tions may be divided into three general groupings: (1) Western Europe 
that is, Switzerland, France, and Belgium; (2) Scandinavia, northern 
Germany, and the eastern shores of the Baltic; (3) The forest belt which 
stretches across northern Russia into Siberia. It is with the first of these 
that we are immediately concerned. 

Commencing with Switzerland, we find, in the so-called Lake Dwelling 
culture of the Neolithic, a blending of the old with the new. The early 
Lake Dwelling culture of western Switzerland, centered about Lake 
Neufchatel, consists of the grafting of North African Neolithic agriculture 
upon a local Mesolithic base, while that of eastern Switzerland represents 
the same phenomenon to which a Danubian element may later have been 
added. Toward the end of the Neolithic period, just before the introduc- 
tion of metal, the Corded people invaded Switzerland from the north, 
and at this time local, sectional differences were to some extent ironed 
out. 66 

Under these circumstances, we may expect to find, in all Swiss Lake 
Dwelling skeletal collections from the Early and Middle Neolithic, exam- 
ples of the small Mediterranean race, representing the bringers of agri- 
culture and animal husbandry to the hunting and fowling communities 
of the lake shores; as well as survivors of the previous population, what- 
ever, in a racial sense, they may have been. 

Unfortunately, the archaeologists have yet to discover the cemeteries 
in which the Lake Dwellers buried their dead; what few remains have 
been found seem to have been for the most part those of persons who died 
by accident, and especially of children. Schlaginhaufen states that seventy- 

66 As suggested by Hooke, Beatrix, G. E., and Morant, G. M., in their article: Bio- 
metrika, vol. 18, 1926, pp. 99-104. 

66 Childe, The Danube in Prehistory, p. 186. 


three Lake Dwelling skulls, suitable for the study of the cranial index, are 
known to exist, but few of these have been made available to the pro- 
fession through publication. Schlaginhaufen could find but nine adult 
crania with measurable faces. 67 

His conclusion is that in the earliest phase of the Swiss Neolithic, 
brachycephals predominate; in the late stages, round and long skull 
forms are about equal in number, with few intermediate forms; later, the 
two blend, and there is a reinforcement of dolichocephals at the begin- 
ning of the metal period. The brachycephals of the Early Neolithic were 
short statured, low faced, low orbitted, and broad nosed; later, their face 
form became longer and narrower, producing, by the end of the Neolithic, 
disharmonic forms. The original combination of round heads with low 
faces and orbits had been upset by mixture with the invading Mediter- 

One must remember that these conclusions on changes in linkage be- 
tween head form and face form are presumably based on no more than 
nine specimens. Five of these may be studied directly from readily avail- 
able published data. 68 Three of them are brachycephalic, two meso- 
cephalic. The former have upper facial indices below 50, and nasal in- 
dices above 50; while the latter fall on the other side in each case. In the 
orbital index two of the brachycephalic crania fall below 80, and one 
above it; while both of the dolichocephals are above. In these five exam- 
ples, then, the round skulls have short faces, low or broad noses, and low 
orbits, while the longer specimens are higher and narrower in face, orbit, 
and nose form. 

The dolicho- and mesocephalic Swiss Lake Dwelling crania seem to 
belong without exception to some variety of small Mediterranean, such 
as might have entered either from the east or the southwest, with agri- 
cultural movements. The brachycephals, which are most numerous and 
least mixed in the earliest levels, form the one element in the Lake Dwell- 
ing racial complex which cannot be derived from known Neolithic sources, 
and may, therefore, be circumstantially linked to the Mesolithic element 
so important in Swiss Lake Dwelling cultures. 

Besides the Lake Dwellings, with their meager supply of human re- 
mains but rich yield of cultural objects which have perished elsewhere, 
there are Neolithic sites of other kinds in Switzerland, including rock shel- 

87 Schlaginhaufen, O., Die menschlichen Skeletr ester aus der Steinzeit des Wauwilersees. 
68 Covering the following crania : 

(1) Pittard, E., ASAG, vol. 7, 1935, pp. 118-122. One fjemale, Lake Neuchatel. 

(2) Pittard, E., Anth, vol. 10, 1899, pp. 281-289. One female, Point, Lake Neuchatel. 

(3) Schenk, A., REAP, vol. 15, 1905, pp. 389-407. One female, Lake Leman. 

(4) Kollman, J., KDGA, vol. 29, 1899, p. 116. One female, Auvcrnier. 

(5) Schlaginhaufen, O., op. at. One female, Greifensee. 


ters and cist graves. Most of these dry land burials, which were not, in 
most cases at least, Lake Dwelling cemeteries, contain human remains 
of Mediterranean type, although a few brachycephals have been found 
in them. 69 

The most extensive single series is that from the cist cemetery of Cham- 
blandes, with ten male and eight female skeletons. 70 (See Appendix I, 
col. 14.) These remains are those of small, light-boned Mediterraneans, 
dolicho- to mesocephalic, mesorrhine, and shallow jawed, with very little 
metrical sex differentiation. Basically, these Chamblandes people resem- 
ble the smaller groups of predynastic Egyptians very closely, but are even 
closer to Muge. There seems to be a perceptible negroid element in the 
Chamblandes groups, which accentuates the African relationship. In 
vault size and height, they do not resemble the Danubians. 

The Chamblandes culture was mid-Neolithic, and probably represents 
the northward intrusion of a semi-nomadic tribe or band from northern 
Italy, where cist burials of the same type have been found. 71 Since the 
Chamblandes physical type is an excellent example of the small Mediter- 
ranean race, that type must, therefore, have been prevalent in the Early 
and Middle Neolithic of northern Italy. Its presence furthermore illus- 
trates the complexity of ethnic movements in Neolithic Europe. 

The racial prpblems exposed by the study of Neolithic man in Switzer- 
land apply equally to France, which presents an even more complex 
archaeological situation. Along the whole Atlantic coast, and most of all 
in Brittany, dolmens and other kinds of megalithic monuments were built 
in abundance. The north of France, especially the Paris Basin, formed the 
westernmost reflection of the Danubian invasions from the east, through 
the mixed cultures of southern Germany, but in the Paris Basin this culture 
was mingled with megalithic elements, since many of the burials are in 
hewn underground vaults and in dolmens. 

The southeast of France contained a surviving cave culture, while the 
whole eastern section of the country, in the valley of the Rh6ne and the 
borders of Switzerland, was occupied by farmers with the same blend 
of Mesolithic and Neolithic cultural elements which in Switzerland ap- 
pear in the western Lake Dwellings. Both Dechelette and Menghin derive 
the agricultural element in the French Neolithic south of the Paris Basin 
from North Africa. 72 

Although, if one may judge by the number of finds made, France was 
a densely populated country during the Neolithic, the distribution of 

69 Schlaginhaufen, O., op. cit. 

70 Schenk, A., REAP, vol. 14, 1904, pp. 335-375. 

71 Childe, The Danube in Prehistory, pp. 163, 174. 

72 Dechelette, J., Manuel d'archaeologie prehistorique. 
Menghin, O., Weltgeschichte der Stein^eit, 


people was very uneven. It is very likely that large areas, notably in the 
Massif Central^ the mountain core of south-central France, where a thin 
soil and granite base are inimical to agriculture, were still inhabited 
throughout the Neolithic time span by scattered bands of Mesolithic 
hunters and grubbers. The bulk of the population lived in the great river 

As an indication of the head form of the French Neolithic people, we 
may turn to a compilation of 608 crania, out of which 43 per cent are 
dolichocephalic, 38 per cent meso-, and 19 per cent brachycephalic. 78 
Although this distribution is not bimodal, there are at least two types 
present, a long and a round one. 

The long-headed type or types belong clearly to the Mediterranean 
category. Although most series include brachycephalic crania, a few are 
purely long headed. Some of them, such as the series from L'Homme Mort 
and Lozre 74 (see Appendix I, col. 15), are low dolichocephals, with means 
of 72; these approach but do not quite approximate the British Long Bar- 
row standards of size. The skulls from the corridor tomb of Vaudancourt, 
Oise, are of full Long Barrow size, and the stature of the skeletons is tall. 
Thus there was, apparently, here and there, a tall, large, and very long- 
headed element in the French Neolithic, related to that which predom- 
inated in the British Isles. It was rarely, however, pure. 

The mesocephalic crania are, as a rule, larger in vault size than most 
of the Mediterranean groups which we have studied, such as the Danu- 
bians, the Chamblandes series, and the Mesolithic skulls from Muge. One 
suspects that the mesocephaly so common among Neolithic French crania 
may, in part, be due to a mixture between a Megalithic, rather than a 
small Mediterranean, dolichocephalic type with brachycephals. This is 
supported by the evidence of stature, for means of French Neolithic series 
run to 164 and 165 cm., taller than the majority of Mediterraneans 

In certain definite ways, the long-headed crania of the French Neolithic, 
as a whole, show a western affiliation: the vaults are wider than they are 
high, and the noses are leptorrhine or low mesorrhine. In these respects 
they differ from the Danubians, as well as in size; and in the vault form, 
they differ from the Corded group. These peculiarities further strengthen 
the similarity between the longer and larger examples, and the British 
Long Barrow type. We may conclude from this that most of the Mediter- 
ranean racial element in France came from North Africa and the Mediter- 
ranean, and little from central and eastern Europe. 

78 Salmon, P., REAP, vol. 5, 1895, pp. 155-181. Series re-divided to agree with con- 
ventional partitionment of cranial index. 

74 Unpublished measurements by Mrs. Ruth Sawtelle Wallis. 


The geographical distribution of Neolithic crania by head form can be 
partially determined from Salmon's study. 76 In all, forty-one departments 
are represented, covering less than half of France. Of these forty-one, 
only fifteen departments, one-sixth of France, have ten or more crania 
each. As far as we can tell from this fragmentary distribution, there were 
two centers of high brachycephaly, one in the Auvergne region, crossing 
the Rh6ne to Savoie, and fading out in the Massif Central; the second in 
the north of France, from Paris over to the Meuse. The Atlantic coastal 
region below Brittany, and the west central part of France, were dolicho- 
cephalic strongholds during the Neolithic. 

The range of indices in the French Neolithic extends from 63 to 97, 
which is practically the normal range for the world. Whole groups of 
over thirty skulls (as at Beaumes Chaudes), found in single caves, are 
entirely long headed, showing that some purely dolichocephalic local pop- 
ulations existed in Neolithic France, as they do in parts of the country 
today 76 (see Appendix I, col. 16); while smaller interments contain wholly 
brachycephalic clusters. Hyperbrachycephaly had already developed as 
an evolutionary phenomenon, for twenty-five out of Salmon's six hundred 
and eight crania have indices between 85 and 97. Others over 90 were 
found in the Swiss collection. This extreme head form was not, appar- 
ently, as common then as it is today. 

Salmon's list luckily contains data as to mode of interment as well as 
to cranial index and locality. Most of the crania come from either mega- 
lithic tombs or caves. Rock shelters and caves contain the same head form 
ratio as the total for France; and this is also true of the totality of mega- 
lithic tombs. Brachycephalic crania are found in all kinds of interments; 
there is nothing of an archaeological nature to distinguish them socially or 
ethnically from the others. They were, therefore, an integral part of the 
Neolithic population in all sections where they have been found. They 
cannot have belonged to a separate, unified group of immigrants, but 
formed rather a residual element in the total population, with a strong 
genetic impulse for the perpetuation and increase of its peculiar head form, 
regardless of other racial factors. 

The further examination of this problem of western European brachy- 
cephaly can best be pursued by a study of Belgium, which formed an 
extension of the archaeological province of northern France during the 
Neolithic. Most of the sites of this period come from the Ardennes hills, 
from the present Walloon-speaking part of Belgium, for the swamps and 
fens of the Flemish country offered little inducement to Neolithic farmers. 

76 Salmon, op. cit. 

76 Bonin, G. von, considers the Beaumes Chaudes series a Palaeolithic survival into 
Neolithic times. HB, vol. 7, 1935, pp. 216-217. 


It is perhaps for this reason that Neolithic Belgians were even more 
brachycephalic than their relatives in France out of seventy skulls of 
both sexes, 77 one-half have cranial indices above 80. The largest series, 
that of Hastiere, 78 has a mean of 79.8; and a high variability. 79 

Among the readily available published crania one may senate eighteen 
male specimens 80 for which adequate measurements have been given. 
The eighteen adult male skulls divide themselves naturally into two sub- 
groups, of eight and ten, respectively. The first ranges in cranial index 
from 74 to 77; the second from 80 to 83. This natural division is so marked 
that it would be futile to seriate the eighteen as a whole, for the mean would 
fall at a point unrepresented by a single specimen. Seven female crania 
which accompany this series likewise have none in the middle brackets. 

The dolichocephalic group of eight male skulls belongs to a normal, 
Mediterranean type, mesocephalic, and relatively low vaulted. The 
brachycephals (see Appendix I, col. 17), the important group for our pres- 
ent purpose, may serve, through comparison, to help elucidate the prob- 
lem of western European Neolithic brachycephaly. 

In Switzerland we had only a few individual crania for study; in France 
the brachycephalic crania are mingled in individual series with dolicho- 
cephalic ones. In the small Belgian group of ten males, however, we have 
a purely brachycephalic series for comparative purposes. 81 

In searching for the prototype of these Neolithic Alpine skulls, one turns 
naturally to the few Upper Palaeolithic and Mesolithic crania of brachy- 
cephalic type for comparison. 82 In vault diameters, the Neolithic skulls 
correspond nearly to the Ofnet ones of the same sexes, but the female 
examples are smaller than those from the Upper Palaeolithic. All speci- 
mens, of all three periods, are low vaulted. 

The Neolithic Alpine faces, insofar as we can judge, run somewhat 
smaller and narrower than do most of the earlier ones; the orbits are 
much the same, but the noses seem smaller in size. On the whole, the 
Neolithic brachycephalic crania are less rugged and much smoother than 
the earlier examples, more globular, and more infantile. The faces look, 
in many cases, little different from those of the Mediterraneans which 
accompany them. The stature of these brachycephals varies, but is greater 
than that of the accompanying long-heads, reaching 165 or 166 cm. for 

77 Including those on Salmon's list and others. 

78 Salmon, 1895, 33 crania. 

78 Range = 72-88, <r = 3.65. 

80 Anvers, 3; Sandron, 10; Pr6alle, 4; Grotte du Docteur, Huccorgen, 1. 

81 These crania come from the same series as the dolicho- dhes, Sandron and Pralle, 
plus the Huccorgne cranium. There is no such thing as an exclusively brachycephalic 
Neolithic group of any size from any one place. 

* Solutr6 #2, #5, Le Placard (1881). Solutr6 #1 and #3, and Le Placard F and B, are 
high mesocephals. Among Mesolithic crania, Ofnet 1800, 1801, 1802, 1806, 1815. 


males in the few ascertainable instances. This correlation would favor the 
Upper Palaeolithic comparison. 

It is impossible, in an orderly and logical manner, to explain the pres- 
ence of these ancestral Alpines during the Neolithic in Europe west of the 
Alps, north of the Pyrenees, and south of the Rhine. But certain hypothe- 
ses 83 merit discussion, and by elimination of lesser probabilities we may 
narrow the field. The most important of these hypotheses are: 

(1) The Alpine brachycephals came into the area in question during the Neolithic 
period, as part of an agricultural invasion, from the east. This theory, which 
has been accepted as fact by the majority of anthropologists for some 
thirty or more years, may be practically ruled out. All the evidence in 
existence serves to contradict it. 

(2) The Alpine brachycephals came into the area in question during the Meso- 
lithic, as part of a preagricultural invasion, from North Africa by way of the 
Iberian Peninsula. This theory is based upon the discovery of allegedly 
brachycephalic crania at Muge in Portugal. Vallois has recently shown 
that the Muge crania are in reality of Mediterranean type, and that most, 
if not all, of the alleged brachycephaly was due to the post-mortem de- 
formation of a few skulls. Hence, in its usual form, this theory may also 
be considered unlikely, although less improbable than the first. 

(3) The Alpine brachycephals are Afalou type round- heads, carried up to western 
Europe with the Mesolithic movements from North Africa, or from Asia by some 
unknown Mesolithic movement. We have already suggested that the Ofnet 
skulls might have had some such origin. But the Alpine crania are smaller, 
and more globular. The faces are much smaller, though similar in propor- 
tions. These differences may possibly be explained by mixture with small- 
headed and short-statured Mediterraneans. 

(4) The Alpine brachycephals represent a continuation of the Aurignacian brachy- 
cephalic tendency found at Solutre. The Azilian culture was a blend of 
Capsian and Magdalenian elements. It is possible that a brachycephalic 
element from Palaeolithic France passed into this Mesolithic cultural 
expression, and was carried over into the Neolithic, which retained many 
Mesolithic cultural forms. 

(5) The Alpine brachycephals are the result of a genetic tendency toward a globu- 
lar skull form acting on a dolichocephalic group. Without reasonable doubt, 
there has been a tendency toward an increase in brachycephaly in the 
Alpine racial zone in modern times. We are as yet unaware of its true 
cause and of its mechanism. But we cannot, for various reasons, suppose 
that the Neolithic Alpines were merely brachycephalized Mediterraneans. 
They were often taller, and had larger vaults, lower orbits, shorter faces, 

83 The hypothesis that they were the ancestors of the Lapps serves in no direct way to 
explain their origin, and will be dealt with later. 


and wider noses. Furthermore, the soft-parts of living representatives of 
this type are distinctly un-Mediterranean. 

The true answer to the question, "What is the origin of the western 
European Alpines?" cannot yet be given, But we may be reasonably 
certain that they are older than the Neolithic, that they may owe part, at 
least, of their reduced size of vault and face to mixture with Mediter- 
raneans, and that their round headedness possessed a strong genetic sur- 
vival value. At the moment, the theory of an Upper Palaeolithic survival, 
somewhat reduced in head and face size, seems the most reasonable. 


Let us next move to the center of the second area of maximum Meso- 
lithic survival southern Scandinavia. Here the Neolithic cultures and 
techniques were late in arrival, and survived long enough to attain a 
considerable complexity, flourishing long after most of the rest of Europe 
was making common use of metal. The old Erteb^lle country of Denmark 
and southwestern Sweden became the seat of a dense population of suc- 
cessful farmers and cattle breeders, partly derived from the old fishing 
and hunting stock, and partly from new immigrants who brought with 
them new ways of living. This part of Scandinavia, in the Sub-Boreal 
period, which followed the Litorina, and which witnessed the develop- 
ment of the Neolithic, was eminently suited to agriculture and cattle 
raising, for the climate was drier than at present, and four Fahrenheit 
degrees warmer in mean annual temperature. 84 

Neolithic impulses, when they eventually reached Scandinavia, prob- 
ably no earlier than 2500 B.C., came into this region from more than one 
direction. It is possible that Danubian influences, transferred through 
South German mediums, were felt by the Ertebjzflle moor-dwellers at the 
beginning, and also that Neolithic cultural movements came directly to 
Scandinavia from South Russia. However, the first movement which can 
be traced with certainty was that of the Megalithic immigrants. These 
came by sea from the south and west, probably for the most part from the 
British Isles, although some may have come from Brittany as well. They 
brought with them not only the habit of erecting impressive burial monu- 
ments, but also agriculture and animal husbandry, which they may have 
been the first to introduce as a basic source of food supply, although 
Neolithic techniques may have come from the east and south before them. 

The Megalithic invaders found a strong, settled population of fishermen 
and hunters, located mostly on the coasts, who apparently did not prevent 
them from establishing their farms and trading stations. The archaeo- 

84 Shetelig, Falk, and Gordon, Scandinavian Archaeology, p. 53. Much of this intro- 
ductory material is based on their book. 


logical record furthermore makes it certain that the aborigines were 
neither driven out nor destroyed, but survived to form an important ele- 
ment in the eventual Danish population. 

The forms of the abundant megalithic monuments, in combination 
with weapon types, provide a scale for Neolithic chronology. After a 
tombless period characterized by round-poled axes, dolmens were built 
first, followed by passage graves, under specific influence from Brittany 
via Holland; and by Long Barrows brought, as a trait, from England by 

In the later part of the dolmen period and the beginning of the passage 
grave epoch, a new group invaded Scandinavia from the east and south- 
east, probably initially attracted by the rich supply of amber in Jutland. 
These were the so-called Battle-Axe people, who were simply our old 
friends the Corded people under their alternate name. Their route lay 
from Holstein up through Schleswig to Jutland, and only later did they 
reach the Danish archipelago, and Sweden. Having come from Germany, 
it is doubtful if they represented a pure Corded racial strain; this became 
less pure through blending with their predecessors in Scandinavia, the 
Megalithic and Kitchen-Midden peoples. The burial form of the resultant 
amalgam was the stone cist, a Megalithic-Corded compromise, with the 
corridor tombs and Battle-Axe single graves as prototypes. 

During the entire Neolithic, almost all of Norway, as well as central and 
northern Sweden, remained in a food-gathering stage of culture, although 
Neolithic axes and other objects were traded to them from the south. 
There can be little doubt that to. a large extent the northern hunters were 
direct descendants of Mesolithic, and hence of Late Palaeolithic, man. 
Many traits of their so-called Arctic culture have survived until recent 

Without the knowledge of Neolithic movements and continuities pro- 
vided by the careful work of the Scandinavian archaeologists, and with- 
out a previous study of the Neolithic racial situation in other parts of 
Europe, it would be difficult to interpret the human remains from the 
Danish and Swedish sites, since this is racially the most complex and most 
mixed section of the continent. The concept of Scandinavia as the home 
of a pure Nordic race or of any other single group during the Neolithic 
is a completely false one. 

The total of Neolithic skulls from Scandinavia is well over two hun- 
dred; 86 of these nearly three-fourths come from Denmark. Only one repre- 

85 Principal sources : 

Fiirst, G. M., %ur Kraniologie der Schtvedischen Stein&it. 
Nielsen, H. A., ANOH, 1905, 1911, 1915. 
Retzius, A,, Crania Smcica. 


sents Norway, and this is a heavy-boned specimen, with strong browridges, 
a mesocephalic vault, mesorrhine nose, and low orbits; apparently a par- 
tial or complete Mesolithic survival. 

In both the Swedish and Danish series, two main, mutually contrasting 
types are found. One is a very long, quite narrow, cranium of moderate 
height; with projecting occiput, parallel side walls, moderate browridges, 
a moderately sloping forehead, which is usually quite broad; a moderate 
upper face height coupled with a narrow breadth; mesoconch orbits of 
square form sloping downward at the outer corners; and a mesorrhine or 
leptorrhine nasal aperture. This type of skull, which comprises some 
thirty-nine per cent of the Swedish series, and five per cent of the Danish, 
was early recognized by Fiirst as a counterpart of the British Long Barrow 
race, which occurs more frequently in Britain in unmixed form. In the 
Danish Long Barrow tombs of purely British type, the skull form is also iden- 
tically British. 86 Most of the people of this type in Neolithic Scandinavia 
must have come by the western sea route around Britain; some, however, 
may have arrived overland from southern Russia in pre-Corded times. 

This Megalithic form is not, however, the only long-headed type dis- 
cernible among Scandinavian Neolithic long-heads; individual crania of 
Corded type with longer faces and higher vaults are not uncommon. A 
mean stature of 172 cm. for the long-headed skeletons 87 shows that the 
racial types involved were tall, taller than either the Long Barrow mean 
from England or that of the Corded group from Silesia and Bohemia. But 
this excess of stature cannot be taken to indicate a strong admixture in 
this type of Palaeolithic long heads, for the dimensions of the vault are 
not comparable, and the face is very narrow as with both Megalithic and 
Corded crania elsewhere. 

Unfortunately, it is impossible to follow the progress of these long heads 
through the different types and stages of Neolithic cultural development. 
Dolmen burials and those in corridor tombs have been classed together 
in Denmark and may be contrasted with profit only with the skeletons 
from the later cist graves. In both groups there has been much mixture 
between long- and round-headed forms; a mean cranial index of 77 in 
each case indicates an intermediate condition. Since the brachycephalic 
element in each is probably the same, and apparently present in equal 
quantities, we may compare the two groups with some validity. The cist- 
grave crania are higher vaulted, longer and generally larger faced, and 
longer nosed than the Megalithic ones. In all diverging characters, the 
cist grave skulls differ from their Megalithic predecessors in a Corded di- 

86 Five crania from Danish Long Barrows. 

87 Pearson's formula, M ~ 172.4 cm. Nielsen's figure is 173.4cm., based on Ma- 
nouvrier's tables. 


rection. Therefore, we are led to believe that a true Corded racial ele- 
ment did play a perceptible part in the formation of the Neolithic Danish 
population, and did not appear merely as sporadic individual specimens. 

In Sweden, out of twenty-four male crania found in passage graves, 
only one was brachycephalic; for the most part a pure Long Barrow type 
is represented. 88 In the later cist graves, a much stronger brachycephalic 
element had entered. On the whole, the Swedish material runs more 
strongly to both extremes than that from Denmark (see Appendix I, 
cols. 18, 19); forty-nine per cent of the Swedish skulls are considered mix- 
tures between the long- and round-headed forms; while in Denmark these 
total eighty-seven per cent. In Sweden, the round heads are concentrated 
in Skane, in the southwestern part of the country; in Denmark, they are 
commonest on the islands of Zealand, Laaland, and Falster. The long- 
heads were particularly prevalent in central Sweden and in Jutland and 
the islands of Fiinen and Langeland. Brachycephaly, therefore, is centered 
around the Copenhagen region, and particularly the islands, which would 
naturally permit the greatest survival of people who derived their suste- 
nance from the sea. 

From every standpoint it seems indicated that this brachycephalic ele- 
ment in the population is associated with the preagricultural midden 
dwellers. Yet we know from our scanty list of Mesolithic remains that the 
basic element of that time was probably a long-headed, Brunn-like Up- 
per Palaeolithic European survival. Many skulls of large, square-jawed 
brachycephalic type appeared toward the end of the Mesolithic or begin- 
ning of the Neolithic in Denmark and northern Germany. Most of them 
have been assigned, largely through caution, to the Neolithic rather than 
to the preceding food-gathering period. Such are the skulls from Kiel, 
from Plau, from Spandau, and numerous other sites. 89 

Whatever their date, they resemble the brachycephalic crania of un- 
disputed Neolithic age very closely. The latter, in turn, are sufficiently 
numerous for accurate racial evaluation. The Danish and Swedish brachy- 
cephalic people were tall, with a mean of 168.2 cm., 90 and heavy boned. 
Their skulls are large, high vaulted, and with lengths greater than those 
common to most crania of equal index. The browridges are usually heavy, 
the foreheads often sloping, the lambdoid region is flattened often, the 
occipital region more rarely. The face is short and wide; the orbits square 
and moderately low; the nasal skeleton often prominent; the nasal index 

88 Furst, op. cit. 

Retzius, op. cit. 

88 Aichcl, O,, Der deutsche Mensch. 

Clarke, J. G., The Mesolithic Age in Northern Europe. 

Kossinna, Gustav, Ursprung und Verbreitung der Germanen, MannusB, #6a, 1928, 

90 Pearson's formula, 170.7 cm. by Manouvrier's tables, 


usually leptorrhine or mesorrhine; the lower jaw heavy, wide, and angu- 
lar. There seems little reason to dispute the conclusion that this type of 
skull is closely related to that found at Ofnet, Bavaria, in the Mesolithic; 
and that it is at least strikingly similar to the Upper Palaeolithic brachy- 
cephals from Afalou bou Rummel in Algeria, to which the Ofnet crania 
have already been compared. Individual Scandinavian crania can be 
matched with others from Afalou. 

Brachycephalic crania are not infrequent in the Neolithic graves of 
central and southern Germany, in which we have already found them 
mixed with long-headed varieties. The same is also true of Poland. In the 
southwest, the Danish brachycephalic type, commonly given the name of 
the site Borreby, is found as far from its apparent center as Belgium, where 
the three crania of Sclaigneux are probably marginal representatives. 91 
In the absence of further knowledge, one cannot definitely state that this 
brachycephalic type was the principal one of the Erteb011e kitchen-midden 
period, or that it was not. But it seems most reasonable to suppose that it 
was native to southern Germany during most of the Mesolithic, with ex- 
tensions westward and eastward; and that at some time during the Late 
Mesolithic or initial Neolithic it filtered into northern Germany and the 
coastal zone from Belgium to Denmark and southern Sweden where it 
survived the Megalithic and Corded invasions, and where it is still present 

It is interesting that in the whole stretch of the European continent in 
which Neolithic invaders blended culturally with the previous Mesolithic 
population, from southern France to Sweden, some form of brachycephal 
should appear. This northern Borreby type is different from the Alpine of 
France, Switzerland, and Belgium in a number of ways. The vaults are 
higher, the orbits somewhat lower, the faces larger, the jaws heavier. 
Whereas the French crania are usually globular, many of the Borreby 
ones resemble modern planoccipital types in angularity of vault form. 
The Borreby people, while shorter than their longer-headed companions, 
were quite tall; the Alpines, frequently taller than theirs, were shorter 
than the northern brachycephals. One is tempted to interpret the differ- 
ence partly in terms of the types with which each mixed; a Megalithic and 
Corded mixture with an Upper Palaeolithic brachycephalic type would 
have a quite different result from that of a Danubian or Spanish small 
Mediterranean strain with the latter. In either case, we still may ask: 
What became of the long-headed Palaeolithic element which accompanied 
the brachycephals both in western Europe and northern Africa? 

But this problem is far from solution; we have established the presence 

* Virchow, R., AFA, vol. 6, 1873, pp. 85-118. In Virchow's article skull #3 is the 
subject of a misprint. The length should read 175 mm., the breadth 151 mm. 


of brachycephals in the earliest Neolithic horizons in various parts of 
western Europe, in each case in connection with a strong Mesolithic cul- 
tural survival. We must await further evidence from the mysterious 
Mesolithic for an answer. 


From the Baltic to the Urals stretches a belt of forests and swamps, 
crossed by many rivers, which long formed a shelter for primitive hunters 
and fishers, while the steppes to the south were overrun by successive 
groups of farmers and pastoral nomads from the earliest Neolithic until 
modern historical times. This northern cultural backwater forms environ- 
mentally a westward extension of the vast Siberian expanse of tundra and 
taiga; since early pre-Slavic days it has been the home of various tribes of 
Finns, some of whom once led, on European soil, a life much like that of 
the Siberian Ostiaks and Voguls of recent centuries. 

In the Neolithic time-expanse, in the general European sense, the in- 
habitants of these forests lived by hunting and stream-fishing, in a manner 
reminiscent of their Maglemose predecessors. A few cultural innovations 
filtered northward from the agricultural lands, and among these was pot- 
tery, decorated by comb-impressions and other characteristic marks which 
render it easy to identify. Within the last few years there has been much 
discussion about this combed pottery, for it has been found in a more or 
less continuous band from Finland across Russia into Siberia, and then 
again at various points across the northern forest region of North America 
to the Atlantic. A school is rapidly forming which believes that this type 
is circumpolar and boreal, non-agricultural, and associated with the 
hunting and fishing peoples of the entire north. An impressive roster of 
archaeological authorities, including Kossina, Ailio, and Childe, believes 
that in Europe it was associated with an early Finno-Ugrian forest people, 
the direct ancestors of the various Finnish groups of today. 92 

The skeletal evidence from the Neolithic of this forest belt, while not 
abundant, is sufficient to show that racial uniformity did not characterize 
this widespread cultural province. Fifteen crania from the Neolithic of 
the shores of Lake Ladoga 93 are almost equally divided into two types; a 
normal South Russian dolichocephal, presumably of the extreme long- 
headed type, with narrow face and nose; and a mesocephal which does 
indeed have a Finnish appearance in the modern sense. Skulls of the latter 
type are characterized by low orbits, short, broad noses, and wide faces, 
which as individual examples exceed the accompanying brain case in 
width. This face and head form bears a certain Cr6-Magnon-like 

92 Childe, V. G. } "Adaptation to the Postglacial forest on the North Eurasiatic Plain," 
in McCurdy, G. G., Early Man. 

* Bogdanov, A. P., 1882; from Sailer, K., AAnz, 1925. 


implication, and may indeed indicate descent from some eastern Upper 
Palaeolithic form as yet undiscovered. 

At Salis Roje, in Livonia on the Gulf of Riga, another collection of 
thirty-one Neolithic crania is even more varied. 94 This includes not only 
the types present at Lake Ladoga, but also a short-statured, brachycephalic 
form, with a long face, slight prognathism, high orbits, and a broad nose. 
Morphologically, there is said to be a mongoloid appearance to these 
crania. This adds, therefore, a third element to the northern forest popu- 
lation during the Neolithic. 

Farther to the east, at Volosovo on the bank of the Oka River, a sub- 
brachycephalic skull from the same cultural horizon would apparently fit 
into the Finn-like Ladogan category. 95 Across the Urals in Siberia, the 
essentially European character of the Comb-Pottery people comes grad- 
ually to an end. A female skull from Bazaiha 96 in the Krasnoyarsk district 
resembles the Salis Roje brachycephalic type, but has a narrow, prominent 
nose. This specimen has been likened to a form typical of modern Turko- 
Tartar women. Farther to the east, one encounters a hyperbrachyce- 
phalic, fully mongoloid skull from Kokui on the Transbaikal railroad, 97 and 
beyond that the extensive and carefully studied Neolithic series from Lake 
Baikal, the main type of which Debetz finds identical with the crania of 
modern Tungus. 98 

In summarizing this material, we shall not dispute the opinion of the 
archaeologists who have concerned themselves with this special field that 
the participants in the comb-ceramic hunting and fishing culture of 
northern Russia and the forests to either side were the cultural ancestors 
of some, at least, of the modern Finno-Ugrian-speaking peoples. But the 
racial aspect of the problem is far from simple; at least three elements were 
present; an extremely long-headed Mediterranean form with southern 
connections; a Cr6-Magnon-like broad-faced, low-orbitted mesocephal, 
filling most closely the requirements of an ideal modern Finnish type; and 
a small-statured brachycephal with a long face and high orbits, which in 
some instances is at least partly mongoloid. As will be seen later, the sub- 
brachycephalic element in the Danubian population was probably related 
to these non-Mediterranean forest types. 


The survey of the white race during Neolithic times, which has required 
the wholesale examination of a large number of skeletal remains and their 

* Virchow, R., ZFE, vol. 9, 1877, p. 412. Also, Sailer, K V} AAnz, 1925. 
96 Pavlov, A., RAJ, vol. 16, 1927, p. 56. See also Ouvarov, A. S., Archaeologie de la 

w Dus, AF, vol. 1, 1923, pp. 72-78. Also, Sailer, K., AAnz, 1925. 

w Dus, ibid. Sailer, ibid. M Debetz, G,, RAJ, vol. 19, 1930, pp. 7-50. 


placing in space, time, and cultural settings, has led to a number of definite 
conclusions, some of which are as final as anything can be in the present 
state of physical anthropology, and others which are admittedly both 
tentative and tenuous. 

The Neolithic manner of living differs radically from that of Palaeolithic 
and Mesolithic man, since it involves the production of food by agriculture 
and animal husbandry. The plants and animals themselves are not of 
European origin, but are native for the most part to western Asia. Neo- 
lithic civilization had probably begun in Egypt, Mesopotamia, and pos- 
sibly the Indus Valley by 5000 B.C. The people who discovered or in- 
vented this control over nature probably belonged to the purely sapiens 
branch of the white race in the larger sense, including a group of related 
dolicho- or mesocephalic types which did not form part of the more spe- 
cialized European and North African Upper Palaeolithic group, although 
they were closely related to such generalized forms as Galley Hill and 
Combe Capelle. 

Members of this larger racial group invaded Europe from several quar- 
ters, starting in the latter part of the fourth millennium B.C. Their principal 
avenues of approach were from North Africa through Spain, from the 
Mediterranean to western Europe by sea, across the South Russian plains, 
and up the Danube Valley. The Danubian migration may have been fed 
by streams from north of the Black Sea, from Anatolia by way of the 
Bosporus, from southern Anatolia and points farther south and east by 
way of Greece, or by some combination of these three. The exact source 
or sources of the Danubian migration remain to be determined. Another 
avenue was to Greece and Italy from the east by sea. 

The invaders may be divided into a number of sub-types. First, there is 
a basic cleavage into a short-statured, sexually undifferentiated, relatively 
small-headed and frequently mesocephalic variety which fits most closely 
the specifications of the Mediterranean race in the more commonly used 
sense of that term. There were three groups of Neolithic culture bearers 
who belonged principally if not entirely io this type: the Danubians; the 
farmers and swineherds who moved westward along the fertile coastal 
regions of North Africa, and over into Spain and thence northward to 
France and Switzerland; and the sea-borne settlers of Italy, and probably 
also of Greece. The Danubians are distinguished by a particularly high 
cranial vault and high nasal index; the western branch by a lower vault 
and narrower nose. To the latter class belonged also the ancient 

The other half of the Neolithic Mediterranean race is noted for tall 
stature and a more extremely dolichocephalic skull form. This variety 
was found in East Africa; it was also common in early Mesopotamia and 


Iran, while the Egyptians belonged more nearly to the smaller Mediter- 
ranean variety. This tall, longer-headed half of the race is longer faced, 
narrower nosed, and less delicate in bony structure than the other. It also 
seems to fall closer to such possible prototypes as Galley Hill and Combe 
Capelle from the Palaeolithic. 

This tall branch is again sub-divided. One sub-branch, with moderate 
vault and face heights, travelled, in all likelihood, by sea from the eastern 
Mediterranean to Gibraltar, around Spain, and up to western France, 
Britain, and Scandinavia. In the last two countries, and especially in the 
British Isles, it contributed an important element to the population. It is 
not easy to find the prototype of this Megalithic group; some of the Meso- 
potamians seem to have been very close to it metrically, and some East 
Africans as well; we shall later find evidence of it on the shores of the Black 
Sea. For the moment we can only postulate that it came from some as yet 
unidentified part of southwestern Asia, southeastern Europe, or north- 
eastern Africa. 

The other sub-branch, characterized by an extremely high cranial 
vault and a very long face and nose, moved westward from the plains of 
southern Russia and Poland into central and western Europe. The mem- 
bers of this group, who were culturally associated with Corded pottery, 
performed a different part in Neolithic history from that of the Danubians. 
They were not peasants, but traders and presumably warriors. Their final 
destinations were southern and central Germany, especially Saxony and 
Thuringia, and southern Scandinavia. From a late center in the Rhine- 
lands, they were destined to play an important part in subsequent metal 
age prehistory. 

The Neolithic population of Europe did not wholly consist of these 
various invaders just described, although they perhaps made up the more 
numerous element in the whole. In the western and northern fringes, 
away from the gates of entry, earlier peoples of Mesolithic and even 
Palaeolithic tradition remained. In Spain, Portugal, and Italy small 
Mediterranean types of pre-Neolithic or Early Neolithic dating may well 
have blended with the invaders in large numbers, but since the two ele- 
ments would have been much the same it is impossible to determine the 
proportions of each. 

In France, Switzerland, and Belgium a major survival of Mesolithic 
cultural factors into the Neolithic is accompanied by a large brachy- 
cephalic increment, which is indubitably related to, and in some degree 
ancestral to, the modern Alpine race. Farther north, from Belgium to 
Sweden and particularly in the Danish archipelago, one finds, under 
similar circumstances of cultural survival, a numerous brachycephalic 
element, called the Borreby type, which is somewhat different from the 


ancestral Alpine form farther south. The northern brachycephals 
were larger headed and definitely higher vaulted and wider faced; with 
taller stature, heavier limb bones, and in many cases heavy browridges, 
wide jaws, and low orbits. The shape of the skull is sometimes angu- 
lar, while that of the Alpines is perhaps more often globular, although 
this difference does not apply to all individuals and should not be over- 

Both the Alpine and the Borreby types bear strong resemblances to the 
few known brachycephalic examples of Upper Palaeolithic crania. The 
Borreby type in particular resembles those from Afalou bou Rummel in 
Algeria. Both also resemble the Mesolithic skulls from Ofnet in Bavaria. 
There can be little doubt that brachycephalic man in western Europe 
was not a Neolithic importation but a Mesolithic survival. It is pos- 
sible that these two types evolved from Palaeolithic man by some process 
which involved the disappearance or absorption of the normal, long- 
headed and numerically more important element. It is also possible 
that they came into Europe during the Mesolithic from some source 
or sources unknown. The Mesolithic is still so much of a blank in the 
racial sense that almost any movement might have taken place without 

Northern Britain, parts of Ireland, Norway, and the north of Sweden 
formed an area of isolation during the entire Neolithic, into which the 
ideas and products of civilization gradually and only partially seeped. We 
do not know, from contemporary evidence, that Palaeolithic man of the 
type already indicated in the same regions during the Mesolithic, survived 
in these spots through the Neolithic, but later evidence will make that 
assumption reasonable. 

The forests of northern Europe east of Scandinavia were inhabited by a 
hunting and fishing people who formed part of a general circumpolar 
cultural group which probably extended with little technical change 
across Siberia to the Pacific, and may have influenced North America. 
In the European and western Siberian segment of this belt, eminent au- 
thority opines on cultural grounds that the Neolithic inhabitants were the 
direct ancestors of an element in the modern Finno-Ugrians physically, 
although not necessarily linguistically. The skeletal remains from this 
region, while few, yet reveal the presence of at least three separate types; 
a presumably Corded variety of Mediterranean; a Palaeolithic-looking 
mesocephal with low orbits and a wide face, which does simulate an ele- 
ment common among the modern Finns; and an incipiently or partially 
mongoloid brachycephal, with high orbits, a long face, and a prominent 
nose, resembling certain modern central Asiatic Turks. 

The racial history of Europe in the Neolithic, therefore, is a problem in 


the balance between new racial streams of relatively uniform type which 
poured in from the south and east, and older, residual elements which sur- 
vived or suffered amalgamation in the west and north. It again reveals 
the marginal character of Europe in the racial as well as cultural sense, 
and shows the necessity of a greater knowledge of race in Asia and in 
Africa if we are to understand our own origins. 

Chapter V 


The dividing line between the Neolithic and the age of metal is difficult 
to draw and essentially artificial. Like that of any other material, the in- 
troduction of copper and bronze into Europe was a gradual process. In 
much of the continent the use of this new substance was first implanted 
on established agricultural peoples, and for this reason it is generally sup- 
posed that the Bronze Age was a period of cultural diffusion but racial 
quiescence. This supposition is only a half-truth. In the areas of high 
civilization, in which metal was first notably used Mesopotamia and 
Egypt the continuity of local branches of the Mediterranean race re- 
mained quite constant in these thickly settled and well-established valleys. 
That this was by no means equally true of the lands to the north and west, 
we shall presently see. 

The Bronze Age was a period of ethnic complexity. It is a unit only in 
the common use of a single metallic alloy by a number of peoples who ob- 
tained the technique of producing tools, weapons, containers, and orna- 
ments of this substance from the lands of earliest civilization. Within its 
span occurred major shiftings of population, if not equalling, at least com- 
parable to those of the Neolithic. 

In the East, where bronze was early and iron late, the Bronze Age lasted 
for fifteen hundred years or more. In Mesopotamia and Egypt the efflores- 
cence of high civilization occurred entirely within the Age of Bronze, and 
by the time that the harder metal had come in, the highest cultural levels 
had long been attained, and the two valleys had lost their cultural leader- 

In Europe, however, bronze furnished in many regions but a brief inter- 
lude of a few hundred years between stone and iron. Only in far pe- 
ripheries, as in Britain, where iron arrived tardily, did the Bronze Age 
flourish long. Here, as in Mesopotamia and Egypt, it lasted nearly fifteen 
hundred years; but the two equal spans barely overlapped. A Neolithic 
child in Denmark might have had a Bronze Age father; similarly a Bronze 
Age child in Britain might have been begotten by a lonely Kelt trained in 
the use of iron and visiting the western islands before his people. 

Most authors make a distinction between the Ages of Copper and of 
Bronze. In both Mesopotamia and Egypt there was an experimental 


. , 







period before the use of tin as an alloy, and the determination of the proper 
proportions of the two metals, were known. Copper spread northward and 
westward in these early days, and many of the weapons and ornaments of 
western Europe in the so-called Chalcolithic or Aeneolithic (Copper Age) 
period resemble early Egyptian or Sumerian forms. The earliest copper 
and bronze objects were carried to outlying and barbarous parts by 
traders, and could only be obtained by those who had something to offer 
in exchange. The Aeneolithic Italian or Spaniard could no more produce 
a metal dagger than a modern Arab can make a machine-gun. In the full 
Bronze Age, however, imported ingots were cast locally into the desired 
form, and there was a smith in every village of consequence. 

During the Neolithic, the farmer or herdsman could shape most of the 
tools and containers which he needed from local materials. Trade was 
carried on more in luxury objects such as sea-shells, than in primary neces- 
sities. But during the Bronze Age, trade affected everyone, for the metal 
with which ordinary tools and weapons were made came from relatively few 
places. Copper came from Spain, the Carpathian region, and the Caucasus. 
Tin was found in Bohemia, Cornwall, and again in Spain. Extensive trade 
necessarily arose to bring the products of these mining regions together. 

In order to possess bronze objects, the European peoples needed some 
valuable commodity to give in exchange. In the north, this was of course 
amber. The principal amber road ran from Denmark to Saxony and 
Thuringia, to Bohemia, to the Inn River in Austria, and over the Brenner 
Pass to the Po. The people of Bohemia acted as middle-men, buying amber 
from the Danes with gold which they had obtained from Transylvania in 
exchange for tin. Thus, even in the Bronze Age, European culture rested 
upon a basis of interchange of local products. 

This extensive trafficking in material objects must have implied con- 
siderable travel on the part of a large class of merchants. Such travel neces- 
sarily meant exchanges of populations in some degree. Childe believes 
that the earliest Bronze Age objects made in central Europe were cast 
by artisans who had emigrated from southern Russia or Asia Minor, 
forming little colonies in the barbarous European villages. 

The Neolithic period in most of Europe fell in a wet, warm climatic age 
during which much of the continent was covered with forest, and this pro- 
fusion of vegetation had hindered migrations and the development of 
pastoral nomadism. During the Bronze Age, however, the Sub-Boreal 
climate, 1 which then prevailed, was more continental and drier; and re- 
gions which had formerly been forested now became parkland, or in many 
cases open steppes. 

1 Much of the Neolithic of Scandinavia, where the Bronze arrived late, fell also in the 


In many parts of the north European plain the drought may have been 
great enough to discourage agriculture and to force some peoples to rely 
wholly on their flocks and herds, thus changing their habit of life from 
farming to pastoral nomadism. Droughts of this kind also fostered tribal 
migrations, and political disturbances in Mesopotamia and Anatolia, in 
the early part of the second millennium B.C., indicate that widespread 
movements of economic origin were prevalent at this time. 

About the middle of the Bronze Age we find the first definite evidence 
of the domestication of the horse as an animal of traction. Horse-using 
nomads invaded Mesopotamia and brought about the Babylonian Dark 
Age. Others, the Hyksos, appeared in Egypt, where they first conquered 
the Delta, and then obtained control over the entire kingdom. In the 
absence of definite information, it has been supposed that these inroads 
were the indirect result of desiccation farther north, where the steppes had 
become too dry for cultivation, and the erstwhile farmers had turned to 
pastoral nomadism. 

Although all movements on the eastern European plain were by no 
means westward, we may find, in later times, significant parallels to the 
Bronze Age migrations which brought the Hyksos to Egypt, the Nasili- 
speakers to Asia Minor, and other barbarians to Mesopotamia. The west- 
ward migrations of the Scyths, Huns, Turks, and Mongols were simply 
consecutive events in a reciprocal sequence which may have commenced 
long before the days of Herodotus. 

All Bronze Age movements were not entirely overland, however. Metal 
seekers from the eastern Mediterranean followed the megalith-builders 
along their sea route from the Aegean to the Italian islands, thence to 
Spain, and around Gibraltar to Britain and the north. During the Late 
Bronze Age movements of peoples may be established archaeologically, 
but the racial interpretation is complicated by the adoption of that un- 
fortunate practice, cremation, which destroys the evidence which physical 
anthropologists require. 


The age of metal began in Egypt and Mesopotamia early in the fourth 
millennium B.C., and by 3000 B.C. it had spread to the Aegean and to 
Anatolia. Crete probably received metal age influences from Palestine and 
Egypt before most of the Anatolian mainland. Cyprus, which bears the 
same name as copper, was another early center, In the diffusion of early 
metal age culture westward along the Mediterranean and northwestward 
up the Danube, the peoples of Asia Minor, Cyprus, Crete, and the Aegean 
played an important r61e, acting as transmitters of impulses which had 
originated in Egypt and Sumeria. 


Let us first examine what Bronze Age skeletal material there is in Asia 
Minor. So far, all of it comes from two sites, Alishar Hiiyiik, which, in its 
later periods, was a Hittite city, and Hissarlik, the seventh level of which 
was Homer's Troy. Both were important centers in the Bronze Age. At 
Alishar, fifty-three skulls have been studied, from seven archaeological 
periods, ranging from the earliest Copper Age, dated from between 2600 
and 2300 B.C., to the Osmanli invasion. 2 

Ten crania from the earliest period (two "Chalcolithic," eight Copper 
Age) are uniformly Danubian in type, both metrically and morphologi- 
cally. The small, high-vaulted, somewhat infantile dolicho- and meso- 
cephalic form, with small face and mesorrhine to chamaerrhine noses, is 
no different from that found at roughly the same time at Anau, at Mariu- 
pol, in the Kiev Government, and in the Danube Valley, in association 
with Neolithic cultures. Two others, which are longer, may belong to a 
Megalithic or Corded variety. The unity of the early food-producing peo- 
ples on both sides of the Caucasus and Black Sea is therefore indicated, and 
from the racial standpoint, the Danubians could have come to central 
Europe from either South Russia or Anatolia, or both. 

In the second and third periods at Alishar, dated between 2300 and 
1500 B.C., and called the Early Bronze Age, brachycephalic skulls ap- 
peared, and these persisted through the period of the Hittite Empire, for 
several centuries after 1500 B.C. The crania are large, low vaulted, and 
only moderately brachycephalic, with lambdoid flattening, and moderate 
browridges. The faces are of medium length, and narrow, although some- 
what broader than those of the earlier Danubian type. The stature of the 
one male observed was tall, 174 cm. 3 

Not all of the Hittite Empire crania are brachycephalic. A long-headed 
variety, which seems to have replaced or outnumbered the brachycephals 
by the time of the Phrygian invasions, is both longer and lower vaulted 
than the Danubian type of the Copper Age; it is characterized by a very 
prominent nasal skeleton of true Near Eastern form, with little nasion 
depression. Bas-relief sculptures of historic Hittites reproduce this hook- 
nosed, open-eyed type of countenance. 

The sequence of racial types in Asia Minor during the metal ages prob- 
ably runs somewhat as follows: the earliest food-producing people were 
the same as those in western Turkestan and southern Russia. The latter 
probably came in earlier times from the highland belt of which Anatolia 

2 Kansu, Shevket Aziz, TAM, vol. 6, #10, 1930, pp. 25-30; ibid., vol. 10, #15-16, 
1934 pp. 105 seq.; BTTK, vol. 1, #1, 1937, pp. 192-202. 

Krogman, W. M., POIC, #20, 1933, app. #4, pp. 123-138; "Cranial Types from 
Alishar Huyiik," in H. H. von der Osten, The Ahshar Huyuk, POIC, #30, Chicago, 
1937, Partiv, pp. 213-293. 

8 Kansu, Shevket Aziz, 1937, Skeleton #3. 



forms a part. Shortly before 2000 B.C., a moderately brachycephalic type, 
with tall stature, entered Anatolia from regions yet to be determined, 
followed by a low-vaulted, hawk-nosed Mediterranean form, which we 
have named "Cappadocian," and which is well known in the present day 
Near East. True Armenoids or Binaries were not, apparently, common 
in early times. 

During the third millennium B.C., the city of Troy, located strategically 
on the eastern shore of the Bosporus, grew from a village to a city, and 
acted as the most important center of diffusion for Bronze Age culture to 
the north and west, especially to the Danube Valley. Troy II, the first 
real city, lasted through much of the 
third millennium, and was razed soon 
after 2000 B.C. The skull of one 
young female . from this level 4 seems 
to represent the same brachycephalic 
type found at Alishar, with which it 
was probably contemporaneous. If 
craftsmen and immigrants were pass- 
ing over the Bosporus at that time, 
carrying metal techniques to central 
Europe, we may, therefore, suppose 
that some of the few round-heads 
found in sites in the Balkans, who 
were at last entering Europe from the 
east, came from this quarter. 

Toward the end of the second mil- 
lennium was built the Ilium which 
the fair-haired Achaeans were later 
to lay waste; and the settlements be- 
tween the important third millennium city and that of Homer's heroes 
were but minor villages. Troy III (Schliemann's sequence), which existed 
through the first century or more of the second millennium, has yielded two 
male and one female skulls. 6 These three belong to one type; a large doli- 
chocephal, with low to medium vault, and a face of moderate size. In gen- 
eral, they resembled the " Eurafrican" type prevalent in Mesopotamia at the 
same time, and the Long Barrow or Megalithic Neolithic form. Homer's 
Troy, which falls wholly within the Bronze Age, is sterile of skeletons. 

In Palestine, at the city-site of Megiddo, twenty-seven skulls have been 
taken from the Copper Age or Chalcolithic level, dated before 3000 B.C., 
and five more from the immediately following Early Bronze Age horizon, 

4 Schliemann, H., Ilios, City and Country of the Trojans, pp. 270-272. 
* Schliemann, H., op. cit,, pp. 509-512. 


After Schafer, H., and Andrae, W., 
Die Kunst des alien Orients, 1925, p. 554. 


which lasted until about 2600 B.C. 6 The crania from both levels are small 
dolichocephals, of a Mediterranean type; they are delicate and feminine 
in aspect, and sexing is difficult. The nose is prominent, with a high root, 
which often springs directly from glabella without nasion depression. Yet 
in many cases a break in the lateral profile is formed by a bulbousness of 
the forehead above glabella. The occipital development is great, and 
prognathism is not uncommon. 

The high-nosed Cappadocian element found in Alishar Hiiyuk from the 
time of the Hittite Empire onward was also, therefore, the prevailing 
racial type of at least one important city of Palestine during the same 
period. Four Bronze Age skulls, two each from the Mount of Olives and 
Am Jebrul, may be included in the same category. 7 One br achy cephalic 
skull, however, has been found in Bronze Age Palestine; in the cave of 
Umm Qatafa, in the Wady Khreitum. 8 This belonged to an adolescent, 
presumably a male, with a vertical forehead, small browridges, and a ver- 
tical occiput. With him was a large, prognathous dolichocephal. These 
two were not buried in the cave, but had been trapped by a fall of rock. 

Returning to Megiddo, we are told that "the skulls from the Hyksos and 
Late Bronze Age burials differ markedly from the Early Bronze and Chal- 
colithic specimens, and altogether appear to form another major physical 
group." 9 What the features of this later group may have been, we cannot 
determine without further information. But we have one other indication 
of racial types in the Bronze Age Near East, and that is the pictures on 
Egyptian monuments, which almost without exception show western 
Asiatics as white-skinned, bearded, and aquiline-nosed. Some are blond, 
but most are brunet. 

After Alishar, our next good series of Near Eastern Bronze Age crania 
comes from Cyprus. The Bronze Age culture which flourished in this 
island is divided into three periods; Early, Middle, and Late Cypriote; 
from 3000-2100, 2100-1600, and 1600-1000 B.C. 10 Three skulls from the 
early period include two brachycephals, which are too fragmentary for 
further study, and one high-vaulted mesocephalic example. In the early 
and middle periods combined, twenty skulls have been studied. Of these, 
forty per cent, mostly from the middle period, are brachycephalic. 11 (See 
Appendix I, col. 20.) The population was clearly mixed, with a long- 

6 Ehgberg, R. M., and Shipton, G. M., Notes on the Chalcolithic and Early Bronze Age 
Pottery of Megiddo, pp. 44-46. 

7 Henckel, K. O., ZFMA, vol. 28, 1930, pp. 238-243. 

s Neuville, R,, and Boureau, R., BSAP, ser. 8, vol. 1, 1930, pp. 33-36. 
9 Engberg and Shipton, p. 46. ' 

NFttrst, C. M., LUA, N. F. Bd. 29/6, 1933. 

11 Furst's 3 EG and 2 MG crania, and Buxton's 15 EC + MC. See Fiirst, op. cit.; 
Buxton, L. D., JEAI, 1920, vol. 50, pp. 183-235; Massari, C., APA, vol. 59, 1929, 
pp. 65-75. 



headed hook-nosed Hittite-like element, and a brachycephalic one. In 
the Late Cypriote period, seventy per cent of forty-seven skulls were 
brachycephalic. The round-headed element was clearly on the increase 
during the Bronze Age, and it may have begun entering the island at any 
time between 3000 and 2100 B.C. Judging from the evidence of Asia 
Minor and Palestine, we may suppose that this took place nearer the late 
than the early date. At the end of the Bronze Age, iron-using invaders re- 
established dolichocephaly. 

The long-headed element in Bronze Age Cyprus was, apparently, the 
typical Cappadocian, or Near Eastern variety of Mediterranean. In the 


FIG. 27 FIG. 28 

Gjerstadt, J.; Lindros, J.; Sjoqvist, E.; Wcstholm, A.; Swedish Cyprus Expedition, 
Stockholm, 1935. Vol. ii, plates CCXVI and CLXXXIX. 

Late Cypriote period, during the prevalence of brachycephaly, an attempt 
was made through artificial deformation to lengthen the head form, pro- 
ducing the so-called "Hittite" style of deformation. In Egypt, Ikhnaton's 
head was similarly deformed, as were those of his two daughters. 

The round-headed element in Cyprus, which appears identical with that 
from Alishar, is numerous enough to warrant statistical comparisons. 
Fiirst calls the skulls Armenoid, and they do resemble Iron Age and 
modern Armenians quite closely in vault size and proportions, but the 
faces and noses of the Cypriotes are smaller in both height and breadth. 


At the same time, they resemble modern Alpines in vault size, while the 
faces are narrower. The bloc of early brachycephals of western Europe 
and North Africa which includes Afalou, Ofnet, and the Borreby skulls 
is quite different, being much larger in vault size and in facial dimensions 
as well. The Cypriotes notably lack the heavy mandible of western Euro- 
pean brachycephals. 

The position of the Cypriotes in the modern racial scheme falls into the 
brachycephalic group of moderate vault size, including Alpines, Arme- 
noids, and Dinarics; the most notable feature is the small face, notable for 
its narrowness, and the light jaw. It is more like the modern Dinarics than 
anything else, since it diverges from the Armenian standard in the same 
way as do modern Albanians. The stature 12 was tall, as with modern 
Dinarics, and the long bones slender. The brachycephalic people who 
entered the Anatolian Eastern Mediterranean region in the latter part 
of the third millennium B.C. were, therefore, ari early form of Dinaric; as, 
one suspects, were the so-called "Armenoids" who came into Mesopo- 
tamia at the same time. This is our first meeting with the Dinaric race. 
Its appearance in western Asia seems quite abrupt, but was probably the re- 
sult of a gradual development, followed by an overflow or evacuation from 
the seat of its characterization. Where this may have been is still unknown. 

Nevertheless, we may eliminate a number of possibilities. It did not 
come from Egypt or Mesopotamia, and it could not have come from the 
northern steppes, which were occupied by dolichocephals. Its place of 
origin was probably not far from Cyprus. Despite the Anatolian evidence, 
it may have developed somewhere in the highlands of Asia Minor or in the 
mountains of Syria, for it is especially numerous in both these places today. 

It could not, presumably, have been an unreduced and unmixed Upper 
Palaeolithic survival. It lacks the size of vault, the width of face, and the 
lowness of orbit which characterize all groups so derived. In face and nose 
form and in size, it resembles the common Mediterranean types of Asia 
Minor, Mesopotamia, and the Irano- Afghan plateau. It may, therefore, 
have been a local and specialized Near Eastern form, brachycephalized 
by some agency and mechanism which will be explained later. 13 From 
its point of dispersal at the eastern end of the Mediterranean, it spread by 
sea into far distant lands. 


The earliest land to receive metal which is considered part of Europe 
was Crete, and there the Bronze Age Minoan civilization began a century 

18 Mean for 13 males =168 cm. 

18 The biological status and origin of the Dinaric race will be explained in the chap- 
ters on the living, particularly in Chapter VIII, Section 6, and Chapter XII, Sec- 
tions 11, 12, and 17. 


or two before 3000 B.C. Crete had been occupied in earlier times by Neo- 
lithic peoples, of whom unfortunately no physical traces remain. The 
Metal Age was introduced by immigrants from two directions from the 
Egyptian Delta, about the time that Menes was extending his power north- 
ward, and from the mainland of Asia, presumably from Palestine. The 
Cretan manner of metal-working was largely of Asiatic rather than of 
Egyptian inspiration. 14 

Although Neolithic remains are absent, the Minoan Age is represented 
by one hundred and more skulls, and a smaller number of long bones, 15 
as well as a considerable body of very realistic fresco painting and sculpture 
in the round. 

The Cretan skulls found at various sites on the island belong to a fairly 
uniform type; this is a small Mediterranean variety with a mean cranial 
index of about 72. Metrically, they could fit perfectly into a number of 
Egyptian collections, from the Naqada predynastic to the Middle Empire. 
On the whole, these Cretan crania are a little smaller, shorter-faced, and 
less leptorrhine than the majority of the Egyptian series, and show leanings 
in the direction of the Copper Age skulls from Alishar, and the Early 
Bronze Age ones from Palestine. The mean type was somewhere between 
Danubian, Cappadocian, and Egyptian forms. 

That this was a short-statured variety of Mediterranean race is shown by 
the long bones; local means vary from 156 to 162 cm. Hence, the Cretans 
were shorter than the Egyptians as well as lower faced. The bodily build 
of the Cretans is well known from fresco painting and sculpture; the local 
ideal of a small waist and wiry, light, but vigorous musculature, which 
occurs so constantly in the Minoan art, must have been based to a large 
extent on reality. Nevertheless, there was a variant minority with broad 
bodies, and, in the women, large breasts; 16 this departure from the usual 
Mediterranean form was also seen in Egypt, and does not necessarily 
imply the presence of an alien race. 

The Minoans were prevailingly brunet in hair and eye color, but in Late 
Minoan times, at least, blondism was known, but apparently not com- 
mon. 17 The skin is represented by Minoan painters as a deep terra cotta 
for men, and white for women. This exaggerates the difference between 
outdoor and indoor habits of life. It again reflects Egyptian influence. The 

i* Ghilde, V. G., The Bronze Age, pp. 19-20. 

16 Evans, Sir A., Palace of Minos at Knossus, vol. 1, pp. 7-13. 

Duckworth, W. L. H., ARBS, vol. 9, 1902-03, pp. 344-355. 

Hawcs, G. H., and H. B., Crete, the Forerunner of Greece, pp. 23-26. 

Luschan, E. von, ZFE, vol. 45, 1913, pp. 307-393. 

Rosinski, B., Kosmos, vol. 50, 1925, pp. 584-637. 

Sergi, G., AJA, second ser., vol. 5, 1901, pp. 315-318. 

M Myres, J. L., Who Were The Greeks? pp. 74-76. 

Myres, J. L., op. cit. 9 pp. 198-199. 


Egyptians, however, rarely colored the wall paintings of their women 
purely white; except in the case of goddesses and such rare mortals as 
Hetep Heres II, the usual color is a pinkish yellow. 

The facial features of the Cretans, if one discounts the conventions of 
the artists, were purely Mediterranean; the straight, prominent nose, with 
its high root, the smooth profile of the forehead, and the lightness of the 
mandible are all clearly shown. The hair form is wavy or lightly curled, 
and the beard, usually clean shaven, was apparently scanty. A variant 
racial type, which may indicate an Alpine element similar to that found in 
Greece (see following section), is seen in a broad-faced fofm, associated 
with a lateral bodily habitus, and an occasional snub nose. Although the 
physical type of the Cretans has changed somewhat since the fall of the 
Minoan power, the features of the happy and athletic people shown on 
the frescoes at Knossus, and the preoccupied frown of the snake goddess, 
are still familiar to us, for they reflect the common heritage of the Mediter- 
ranean race elsewhere. 

Most of the Early Minoan skulls belong to the Mediterranean type just 
described, which shows a blending between the usual Neolithic variety 
and the convex-nosed type prevalent in the Near East. In some sites, as at 
Hagios Nikolas and Patema, the population was exclusively Mediter- 
ranean. In others, a few brachycephalic examples occur, and these ap- 
parently belong to the same type found at Cyprus. 

In the later Minoan periods the brachycephals increased in numbers, 
but never formed more than a minor element in the population, probably 
not more than a sixth at most. Since 70 per cent of the population of Cy- 
prus may have belonged to this type, the Cretans must have kept them- 
selves fairly free from eastern admixture after the initial establishment of 
their national culture and power. At the time of the Dorian invasions, as 
today, the Cretans were still predominantly Mediterranean. 

Toward the end of the Early Minoan period, somewhat before 2100 
B.C., strong Cycladic influences entered Crete, and it is possible that some 
of the Middle and Late Minoan skulls of unusual size and Megalithic con- 
formation may be derived from this movement. The present population of 
Crete belongs largely to a tall Mediterranean type, which may partially 
antedate the Dorian arrival. 18 


The question of the origin of the Greeks has long been an apparently 
insoluble enigma. For centuries, before the development of archaeology 

18 Our data on which is based the assumption that all Cretans were of short stature 
are not numerous. The Philistines, presumably Cretan relatives in Palestine, are 
thought to have been tall, while some of the Mycenaeans in Greece were of large stature. 


as a scientific discipline, history began with Herodotus, and Homer was a 
small window permitting tantalizing glimpses into the most distant past. 
In recent years, however, great advances have been made toward the 
solution of this problem, by the linguistic and historical researches of 
Myres, 19 and by the publication of skeletal material by Fiirst and Kou- 

The historical reconstruction may be briefly summarized as follows: 
During the Neolithic, Greece was culturally connected with North Africa 
and the rest of the Mediterranean basin. The one skull which is known 
is of normal Mediterranean racial type. In the early Metal Age, immi- 
grants from the Cycladic islands, of Asia Minor origin, introduced copper 
to Greece, with the mother goddess cult, and settled on either side of the 
Isthmus of Corinth. In the meanwhile, Painted Pottery people of Danu- 
bian cultural origin came down to Greece from the north, driven by 
Corded people. Thus, by 2000 B.C., there were, from the cultural stand- 
point, three elements in the Greek population: (a) local Neolithic Medi- 
terranean; (b) Danubian from the north; (c) Cycladic people of eventual 
Asia Minor origin. 

Between 2000 B.C. and the period of Homer, Greece was invaded three 
times more; (a) by Corded people (Myres calls them "Kurgan" people), 
who came from the north about 1900 B.C., and who, Myres thinks, 20 may 
have brought the Indo-European basis of Hellenic speech; (b) by Minoans 
from Crete, who founded the "long genealogies"; dynasties of rulers at 
Thebes, Athens, Mycenae, and elsewhere. Most of these entered Greece 
about 1400 B.C., although some may have dated back to 1700 B.C.; (c) by 
"divine born" foreigners, such as Atreus, Pelops, etc., who came from 
across the Aegean in ships, learned Greek, usurped thrones, and married 
the daughters of the kings of Minoan ancestry. 

These foreigners, whom Myres likens to the Normans in English history, 
begat the heroes of the Trojan war. The war itself reflects the close re- 
lationship between these adventurers and Priam's Troy. In the wars, 
the Homeric heroes formed the nuclei of small groups of "companions"; 
these were homeless adventurers, refugees, and poor relatives, who had 
attached themselves to the heroes in a close personal bond. The bulk of 
the Greek army was composed of local conscripts from the various king- 
doms of Greece, who were of a different ethnic origin and who, like 
Thersites, had no especial interest in destroying Troy. 

The post-Homeric and Iron Age Dorians, long regarded as fresh in- 
vaders from the north were, according to Myres's reconstruction, but 

19 Myres, J. L., op. cit., 1930. 

20 In view of evidence to be presented later, it is more likely that the Danubiaiis 
brought it (Chapter VI). 


Greek speakers who had been isolated in the Mt. Olympus region by the 
warlike activities of the Thebans, and who had obtained iron from Asia 

The Greeks of the great period of Athenian civilization were thus the 
product of much mixture from diverse ethnic sources, as the study of the 
origin of the Greek language also reveals. 

The skeletal record can, in part, supplement the evidence of recon- 
structed history. Six skulls from Hagias Kosrnas near Athens represent 
the period of amalgamation of Neolithic Mediterranean, Danubian, and 
Cycladic elements, between 2500 and 2200 B.C. 21 Three are dolicho- 
cephalic, one mesocephalic, and two br achy cephalic. The faces of all are 
narrow, the noses leptorrhine, the orbits high. One may conclude that 
a Cretan type of Mediterranean and the Cypriote Dinaric form were both 

Twenty-five Mid-Helladic crania represent the period after the arrival 
of the Corded or "Kurgan" folk from the north, and during the seizure of 
power by the Minoan conquerors from Crete. 22 Of these, twenty-three 
come from Asine, and two from Mycenae. Needless to say, the population 
of this time was very mixed. Only two skulls are brachycephalic; they 
are both male, and both associated with very short stature. One is of 
medium size, high- vaulted, and narrow-nosed and narrow-faced; the other 
extremely broad-faced and chamaerrhine. They seem to represent two 
different broad-headed types, both of which can probably be found in 
Greece today. 

The long heads are not of uniform type; some, with large vaults and 
strong browridges, with deep nasion depressions, remind one of the larger 
varieties of Neolithic dolichocephals, of both Long Barrow and Corded 
types; and Fiirst feels that a number of them are very similar to the Late 
Neolithic crania from Scandinavia, of about equal age. Needless to say, 
both Corded and Megalithic people were present in Denmark and Sweden 
at about this time. 

The rest of the long-headed crania, which are probably more truly 
representative of the bulk of the Mid-Helladic population, are of the slight- 
browed, high-nosed type familiar in Crete and Asia Minor during the 
same epoch. They, too, are short statured, while the few examples of the 
larger-headed variety are, as is expected, taller. It is impossible, with 
present data, to isolate from the main body of these crania a Danubian 
type, although the latter may well have been present. 

Forty-one Late Helladic skulls, dated between 1500 and 1200 B.C., and 
coming likewise from Argolis, may include those of some of the "divine- 

21 Koumaris, J., RA, vol. 44, 1934, pp. 248-251. 

^Ftirst, C. M., LUA, N. F., vol. 26, #8, 1930; VHPA, vol. 4, 1930, pp. 3-14. 


born" invaders. Among these, one-fifth are brachycephalic, and appar- 
ently largely of the Cypriote Dinaric type. Of the long-headed skulls, a 
large number belongs now to the larger, more heavily marked varieties, 
and fewer to the smaller Mediterranean. The similarity to the northern 
types, and especially to the Corded, is even stronger than before. This 
increase in a non-Minoan direction may perhaps be attributed to the 
arrival of the ancestors of Homer's heroes. 

This survey carries us through the Bronze Age. The racial history of 
Greece in full classical time is not as well documented as that of the periods 
just studied. Until the inception of the slave trade 23 in Athens and other 
centers of manufacture and export, there can, however, have been little 
population change. In Argolis, the Mediterranean racial element is the 
only one clearly shown in six pro to-geometric and "Hellenic" crania. 24 
According to Koumaris's compilation of cranial indices, 25 mesocephaly 
reigned everywhere in Greece during the classical period, and into Hellen- 
istic and Roman times. The mean index for Athens in the great period was 
75.6, on 30 crania. This mesocephaly probably conceals the presence of a 
varied racial amalgam, with Mediterranean strains predominant. The 
Greek colonies in Asia Minor show much the same combination of types 
which we have seen in Greece itself. 26 Mixture with Asiatics must have 
been masked by the essential racial similarity of the populations on either 
side of the Aegean. 

Greek literature and Greek art furnish an abundance of evidence as 
to the pigmentation and the characteristic facial features of the ancient 
inhabitants of Hellas. The Olympian gods, ancestors of the semi- 
divine heroes, were for the most part blond, with ivory skins and golden 
hair. Athene was gray eyed. Poseidon, however, was black haired. 
These gods were little different, if we may believe Homer, from their 
descendants the heroes, most of whom were white limbed and golden 
haired. 27 

Odysseus's herald Eurybates was dark skinned and curly haired; 
Achilles's son Neoptolemos, perhaps by a brunet mother, was rufous. The 
Spartans were said to be blond, and in fifth-century Athens women 
bleached their hair with an herb which turned it golden yellow, in pur- 
suance of a blond ideal. Vase painters of the sixth to fourth centuries were 
able to distinguish blond and brunet color by conventional glazes, and 

23 Zaborowski, S., ARSI for 1912, 1913, pp. 597-608. 
^Furst, G. M./LUA, vol. 26, #8, 1930, pp. 92-95. 
26 Koumaris, J., ACAP, 1931, pp. 218 seq. 

26 Schumacher, O., ZFMA, vol. 25, 1926, pp. 435-463. 
Zaborowski, S., BSAP, ser. 4, vol. 3, 1881, pp. 234-238. 

27 Myres has conclusively demonstrated that the much disputed word f av36s actu- 
ally did mean "yellowish" or "sandy." Pp. 192-194. 


applied this distinction to representations of living models as well as of 

Greek terminology included words for blue and brown eyes, and for 
green ones, the color of an olive leaf, as well; in skin color it recognized 
rosy vascularity, a pallid hue resembling cream cheese or the skin of un- 
ripe apples, a honey color, and a deep brunet. To Phoenician merchants 
and tanned sailors of other nationalities, they gave the name "phoinix," 
comparable to the color of a ripe date, or a bay horse. Thus within the 
Greek commonwealth as without it, all variations of pigmentation known 
to modern Europeans were probably to be found. 

The Minoan convention of a high-rooted nose and a lithe body passed 
over into classical Greece as an artistic ideal, but the portrait busts of in- 
dividuals show that it cannot have been common in life. Villains, com- 
ical characters, satyrs, centaurs, giants, and all unpleasant people and 
those not to be admired, are often shown in sculpture and in vase painting 
as broad-faced, snub-nosed, and heavily bearded. Socrates, who belonged 
to this type, was maliciously compared to a satyr. This type may still be 
found in Greece, and is an ordinary Alpine. In the early skeletal remains 
it is represented by some of the brachycephalic crania. 

On the whole, one is impressed, after looking at the portrait busts of 
Athenians, and the clay masks of Spartans, with their resemblance to pres- 
ent-day western Europeans. This resemblance becomes less marked in the 
art of the Byzantines, however, where modern near Eastern faces are more 
frequent; but the Byzantines lived mostly outside of Greece. As will be 
shown later (Chapter XII, section 14), the modern inhabitants of Greece 
itself differ surprisingly little from their classical predecessors. 


In early Metal Age times influences from Crete and the Aegean, in- 
cluding those from the second city of Troy, spread westward to Sicily, 
Sardinia, Italy, and Spain, reaching also the smaller islands of the western 
Mediterranean. This maritime diffusion was probably carried by sea- 
farers in search of new sources of metal as well as markets for their products, 
and the traders and adventurers followed the old Megalithic routes. In 
the beginning the bringers of metal and the Late Megalithic colonists may 
well have been the same people. 

The evidence of the racial composition of the Copper Age sailors who 
reached Italy and the Italian islands is simple and direct. The moderately 
tall, long-headed, and narrow-nosed Megalithic people who were im- 
planted, during the Late Neolithic, upon the smaller Mediterranean type 
which had preceded them, were followed, during the Aeneolithic, by others 
of the same kind, in the company of equally tall brachycephals. The latter 


resembled the people of the same Dinaric head form in Cyprus, Crete, 
and the Aegean, and without doubt formed a westward extension of the 
same movement. 

In Sicily, which probably received metal earlier than most of the main- 
land or the islands farther west, Copper Age skulls of one series from 
Isnello 28 are all of general Mediterranean type, with the Megalithic 
variety predominant, as shown by excessive skull lengths, moderate vault 
heights, and narrow noses. The mean stature for twenty-four males, pre- 
sumably of this type, was 169 cm. Other Sicilian series, however, do in- 
clude brachycephals, as at Chiusella and Villafratti, with cranial indices 
ranging as high as 9 1. 29 These form, however, no more than one-third of 
the total Aeneolithic series from Sicily. In the true Bronze Age which 
followed, the incidence of these brachycephals increased. 

In Sardinia a large series of sixty-three Copper Age skulls from Anghelu 
Ruju 30 includes sixteen per cent, or ten individuals, of the new brachyce- 
phalic type, while the others resemble the long heads of Sicily, The group 
as a whole, irrespective of head form, was tall. 81 The racial composition 
of Corsica during these periods is known only through the presence of one 
small, short-statured, long-headed female skeleton of either Neolithic or 
Aeneolithic age, and two brachycephalic crania from the Bronze Age. 32 

It would be interesting to supplement this survey of the Italian islands 
with a study of the crania found in the elaborate burial chambers of Malta, 
of late Neolithic or early Metal Age date, but the excavators of these 
vaults, professional and otherwise, literally threw away what was probably 
the longest unified series of human crania ever found, numbering over 
seven thousand. We are told that these early Maltese were "Mediterra- 
neans," and know little else about them. 33 

On the mainland of Italy, Aeneolithic skeletons, which are found mostly 
on the western side of the central portion of the peninsula, belong to the 
same types found on the islands, but brachycephals are more abundant, 
being equal in number to the dolicho- and mesocephals. 84 Some of the 
Aeneolithic Italians of the Campagna and of Latium were very tali and 
large headed, with both mesocephalic and brachycephalic forms. 85 In 

2 8 Giuffrida Ruggeri, V., ASRA, vol. 11, 1905, pp. 56-103. 
Zaborowski, S., BMSA, ser. 5, vol. 6, 1905, pp. 196-199. 

29 Sergi, G., Grant Preistorici della Sicitia; Europa, pp. 270-289. 

80 Sergi, G,, Crani Antichi della Sardegna, 

81 Bruni, E., RDAR, vol. 26, 1924-25, pp. 235-250. 

82 Bloch, A., BSAP, ser. 5, vol. 3, 1902, pp. 333-363. 

83 Tagliaferro, N., Man, vol. 11, 1911, pp. 147-150. 

84 Sabatini, A., RDAR, vol. 29, 1930-32; pp. 577-582. 
Sergi, Europa , loc. cit. 

Mochi, A., APA, vol. 42, 1912, pp, 330-347. 

36 Genna, G. E., PICP, 1932, pp. 60-64; RDAR, vol. 30, 1933-34, pp. 235-262. 


Istria, at the head of the Adriatic, the Dinaric population which is dom- 
inant in that peninsula today had begun to arrive in the Copper and 
Bronze Ages, 36 judging by a series of six female crania which bear definite 
indications of this type, such as flattening of the occiput, narrow face, 
and projecting nasal bones. The new invaders may, therefore, have 
travelled up the Adriatic as well as over the Tyrrhenian Sea. 87 

Reviewing the Italian material, on both metrical and morphological 
grounds we may determine that the round-headed racial type which came 
into the middle Mediterranean with the introduction of metal was of a 
general Dinaric character, and without doubt came from Asia Minor 
and the Aegean, where it first appeared in the last centuries of the third 
millennium B.C. Since the metal ages of the middle and western Mediter- 
ranean were later than those farther east, the chronological aspect of this 
theory presents no contradictions. 

The Balearic Islands, Spain, and Portugal were, of course, the next 
stops in the westward spread of the metal-carrying seafarers through the 
Mediterranean. During the Early Copper Age in Spain, the distinctive 
Bell Beaker culture arose, which was soon to spread northward and east- 
ward into central Europe, and eventually to Britain, as an important 
racial movement; and another culture of equal local importance, that of 
Los Millares in Almeria, developed from eastern beginnings, with an em- 
phasis on the importation of Egyptian and Near Eastern materials, such 
as hippopotamus ivory, ostrich egg shells, and actual Near Eastern 
pottery. 88 The center of Early Bronze Age civilization again lay in Al- 
meria, with el Argar as the principal site, and began about 2000 B.C. 
During this period, which lasted until the Iron Age, there was again 
much Egyptian and Aegean influence. 

Unfortunately, in the Iberian Peninsula, as elsewhere, the human 
record is not sufficient to support the complexity of the cultural. The 
craniologist cannot keep pace with the archaeologist; we cannot, without 
more numerous and more accurately correlated skeletons, tell in all cases 
what physical types went with each archaeological entity. 

In the Balearic Islands, for a beginning, a few dolichocephalic crania, 
and one brachycephal, have been found in the talayots, or corbelled stone 
towers resembling the Sardinian nuraghes and Scottish brochs, which 
were first built in the Copper Age but which were used until the advent of 
iron. 39 Fifty-eight adult and five juvenile crania with long bones from a 

86 Battaglia, R., PICP, 1932, pp. 57-60. 

87 Unless these particular Binaries came overland from Central Europe. 

88 Childe, The Bronze Age, pp. 146-153. 

89 Aranzadi, T. de, BAC, vol. 1, 1923, pp. 134-140. 
Cameron, John, The Skeleton of British Neolithic Man. 
Comas, Juan, Aportaciones al Estudio de la Prehistoria de Menorca. 


naveta> or long barrow, in Menorca, are said to have represented a homo- 
geneous group of people with short stature, long-heads (all cranial indices 
being under 75), low faces, prominent, aquiline noses, and projecting 
chins. The form of the scapulae and humeri of the males showed that 
they had developed great shoulder and arm muscles from slinging, the 
activity from which the islands derived their name. Three other skulls 
from an ossuary at Biniatap are brachycephalic. 40 

In the Copper Age groups from mainland Spain and Portugal, the old 
long-headed types overwhelmingly prevail: out of one hundred and thirty- 
four crania, which represent all that could be assembled for this survey, 
only fifteen, or nine per cent, were brachycephalic. 41 If one includes 
Ariege, Basses Pyrenees, and Aveyron in the south of France, twenty-eight 
crania may be added, of which only two are brachycephalic. 42 One of 
these, from a site near the city of Narbonne, possesses all of the cranial 
and facial features typical of the Bronze Age brachycephals of Cyprus, 
Italy, and the Italian islands. In few of the Spanish instances are extensive 
details given, but it is probable that the brachycephalic crania there are 
also of the same type. 

Many of the dolichocephalic Copper Age skulls are of Megalithic or 
Long Barrow type, while others are of a smaller, less rugged, Mesolithic 
or Neolithic Mediterranean variety. Among the mesocephalic crania, 
some may again be small Mediterraneans, while others, with larger vault 
dimensions, may in many instances be mixtures between Megalithic and 
brachycephalic types. The statures of the large dolichocephalic group 
average about 167 or 168 cm.; taller than most living Spaniards and as 
tall as the Neolithic Long Barrow population in Britain. Other dolicho- 
cephalic crania go with short stature, with a mean of about 160 cm. Un- 
fortunately, it is not possible to determine the approximate proportions 
of Megalithic and Mediterranean types, but the former seem to be at 
least one-half of the total. 

A special development of the Copper Age in Spain was the Bell Beaker 
culture, about which more will be said later, since its chief influence in the 
racial sense fell upon areas in other parts of Europe. It is at present the 
general belief of archaeologists that the Bell Beaker culture arose in central 

40 Cameron, John, PICP, 1932, p. 60. 

41 Aguilo, Juan C., AMSE, vol. 1, 1922, pp. 23-36. 
Aranzadi, T. de, BAC, vol. 3, 1925, pp. 177-206. 

Barras de Aragon, F. de las, AMSE, vol. 12, 1933, pp. 90-123; vol. 9, 1930, pp. 59-64. 

Batista i Roca, J. M., BAC, vol. 1, 1923, pp. 104-133. 

Mendes-Correa, A. A., Os Povos Primitives da Lusitania. 

Tormo, I. Ballester, APL, vol. 1, 1928, pp. 44-53. 

* 2 H616na, Th. and Ph., BAC, vol. 3, 1925, pp. 1-35. 

Lapouge, G. V. de, Anth, vol. 2, 1891, pp. 681-695. 

Vallois, H., Anth, vol. 37, 1927, pp. 277-303, 473-489. 


Spain, shortly before 2000 B.C., from local beginnings. 43 A North African 
origin is rendered unlikely by the supposed absence of a Bronze Age 
south of Gibraltar, although recent work in Morocco has revealed some 
supposedly early metal. 44 Where Bell Beaker burials are found in central 
Europe, the skeletons are almost always of the same tall brachycephalic 
type which we have already studied in the eastern Mediterranean and 
Italy. In Spain, however, they are frequently of the Megalithic race. 
The basis for the belief that the Bell Beaker people of Spain were Binaries 
rests largely upon three cranial fragments from the type site of this culture 
at Ciempozuelos, near Madrid, and upon one complete mesocephalic 
skull from Cerro de Tomillo some forty miles away. 46 

The measurements of the three fragments are uncertain, and their 
allocation to a definite type impossible. 46 However, all three fragments 
appear to be brachycephalic, and one to have a high vault. One has 
strong, another weak, browridges. One seems to have a slight lambdoid 
flattening. In the only fragment which possesses facial bones, the orbits 
are high and the nose narrow. The Cerro de Tomillo skull is not, however, 
a pure dolichocephal, and does resemble, in a partial sense, the Dinaric 
brachycephalic variety which was common in the Mediterranean at 
that time. 

Although there seems to be little doubt in the minds of the archaeologists 
that the Bell Beaker culture developed in Spain, and although eastern 
Mediterranean brachycephals came there at about the same time, the 
manner in which the physical type and the culture became identified with 
each other is still obscure. 

During the Early Bronze Age, after the efflorescence of the Bell Beaker 
people, Spain became a great center of metallurgy and trading activity, 
rivalling the Aegean in importance. The colonists from the east, who 
had originally located themselves in Spain merely as miners and forward- 
ing agents of metal, now settled down to producing the finished products 
of the Bronze Age in Spain itself, for local sale, since disorders in the 
Mycenaean and Minoan realms had apparently cut them off from their 
homelands. 47 Furthermore, the introduction of fresh cultural elements 
from the east suggests that new people had joined them. 

The principal site of the Early Bronze Age, el Argar in the province of 
Almeria, is located near the silver mines of Herrerias, which were worked 
in ancient times. From some thirteen hundred flexed urn burials, seventy 

48 Bosch-Gimpera, P., Real, vol. 4, pp. 345-362. 
"Ruhlman, A., Hesp^ris, vol. 15, 1932, No. 1, pp. 79-119. 

45 Childe, The Danube in Prehistory, Chapter X, pp. 190-201. 

46 Anton, M., BRAH, vol. 30, 1897, pp. 267-283. 
Deslaers, M. H., BRAH, vol. 71, 1917, pp. 18-38. 

47 Childe, The Bronze Age, p. 146. 


skulls have been recovered, of which twenty-nine are those of adult males, 
and forty of adult females, 48 The el Argar series shows quite definitely 
that the Early Bronze Age people of Almeria were not descendants of 
previous inhabitants, but to a large extent a new population, with definite 
Near Eastern relationships, as one might suppose from the cultural in- 

The series as a whole is one of small people, with a mean male stature 
of 1 58 to 1 60 cm. ; the earlier Copper Age immigrants, for the most part, 
were ten centimeters taller. The skulls gravitate around the indices of 76 
and 77; for sixty per cent of male and fifty-eight per cent of female crania 
are mesocephalic. Of the remaining skulls, long heads outnumber round 
heads two to one. The series is not very homogeneous, and the cranial 
index and most other criteria of form show modalities which make it 
certain that the el Argar people included at least two types which had 
not become completely amalgamated. 

The principal cranial element is a normal, rather small variety of Med- 
iterranean, which seems to resemble, both metrically and in description, 
predynastic or early dynastic Egyptian forms, or at the same time, ele- 
ments which entered Spain in the Neolithic. Prominence of the brow- 
ridges at glabella, and a considerable nasion depression, make this type 
of Mediterranean rather unlike the Cappadocian variety common in 
Asia Minor, although metrically there is nothing to prevent such a rela- 

The second type is the new brachycephalic element, which seems to 
have been the dominant one politically, in that two female skulls found 
wearing silver crowns both belonged to it. It was apparently some form 
of Near Eastern brachycephal with which we are already in a general way 
familiar the skull is short, rather than broad; the vault is medium or 
low; the forehead is narrow, the lambdoid region often flattened, while 
the greatest breadth of the vault comes well to the rear. The nose is high 
and narrow, and the nasal bones join the frontal with little depression, 
while a smooth glabella heightens the impression of a high-bridged, Near 
Eastern type of nose. Although the orbits are high and rounded, the face 
is rather low, but the mandible is surprisingly broad, often with everted 
gonial angles. There is also a perceptible amount of alveolar prognathism, 

Although this is not exactly the brachycephalic type which we met in 
the Copper Age, and which became identified with the Bell Beaker people, 
it is, nevertheless, definitely a Near Eastern variety of brachycephal which 
is familiar in Asia Minor and Syria today. The el Argar people represent 
a mixture of elements which could be duplicated in the modern Near 
East, but not one with which, in our ignorance of most of that end of the 

Jacques, V., BSAB, vol. 6, 1887-88, pp. 210-236. 


Mediterranean, we are already familiar. Some of the Mediterranean 
racial contingent may well have been of earlier Spanish derivation, but 
if so the absence of Megalithic and Copper Age forms is surprising. 

In other parts of Spain no such change of population as that of Almeria 
is manifest. Mediterraneans, both large and small, are carried over from 
the Neolithic and Copper Ages, while the larger variety of brachycephal 
also continues, 49 Out in Mallorca and Menorca, the dolichocephalic ele- 
ment seems to remain as the exclusive or predominant one, for the most 
part tall and of Long Barrow vault form. 50 

The westward migrations of peoples from the Aegean and the eastern 
end of the Mediterranean, during the Late Neolithic, the Aeneolithic, and 
the Early Bronze Age, must have affected the populations of Italy, Sicily, 
Sardinia, Corsica, the Balearics, and the Iberian Peninsula to a consid- 
erable degree. These were real colonizations which added new racial 
elements to the Mesolithic and Early Neolithic Mediterranean sub-stratum. 
By the middle of the Bronze Age, the central and western Mediterranean 
lands had assumed the racial characteristics which they still, for the most 
part, bear. Except for northern and central Italy, later migrations were 
to bring little that was new. 


Since the western Mediterranean lands have changed little racially 
since the end of the Bronze Age, it may perhaps be forgiven us if we break 
the continuity of the present chapter, as was done earlier in the cases of 
Mesopotamia, Egypt, and Greece, to discuss, at this point, the origins and 
racial characteristics of certain non-Indo-European-speaking peoples who 
are or were in later times known by specific names the Basques, the 
Phoenicians (as Carthaginians), and the Etruscans. 

In regard to the Basques, it has been observed that the skeletons from 
dolmens of Guipuzcoa, probably of Early Metal Age, resemble those of 
the modern Euskarians of the same province, in stature, in head size and 
form, and in characteristic facial peculiarities. 61 Since the northern shore 
of Spain, in the country occupied by the Basques since the beginning of 
history, is rich in metal ores and was a favorite haunt of Copper and 
Bronze Age sea migrants, it is very likely that a numerically strong western 
Asiatic element, including both Megalithic and Dinaric types, became a 
permanent factor in the local population. When we come to discuss 
the physical anthropology of living Basques, the probability of such an 
influence will be of assistance. 

49 Aranzadi, T. de, Excavacio de Sepulcres Megalitics, pp. 3139. 

Ban-as de Aragon, F. de las, various articles in AMSE, 1921, 1926, 1930. 

60 Barras de Aragon, F. de las, AMSE, vol. 9, 1930, pp. 38-51. 

11 Serra i Vilaro, after Mendes-Correa, 1924. 


The second people, the Phoenicians, who established their principal 
colony at Carthage at the end of the second millennium B.C., and posted 
trading garrisons at various points on the North African coast, both on 
the Mediterranean and Atlantic sides, also settled along the eastern coast 
of Spain, where they founded the city of Cartagena. Except for the Greeks, 
they formed the last of the groups to migrate westward from the eastern 
Mediterranean by sea, but the first to do so in full historical light. 

The physical type of the Phoenicians is well known from the skeletal re- 
mains found in tombs at Carthage. 52 A series of 117 skulls, of which 68 
are male, belong for the most part to one characteristic type; dolicho- to 
mesocephalic, with the cranial index at 75; fairly long vaulted, and hence 
moderately broad; with a very low vault, a moderately broad forehead, a 
short face, high orbits, and a narrow, projecting nose which often springs 
directly from the frontal bone with little or no nasion depression. These 
skulls are in many ways similar to the Megalithic or Long Barrow type of 
the preceding millennium; but, as is to be expected in view of their late 
eastern Mediterranean origin, show modifications toward a shortening 
and widening of the vault, and a beaking of the nose. 

A few related brachycephals, of Dinaric form, are incidental to this 
type, while a number of less characteristic skulls, with lower orbits and 
less prominent, wider noses, may be those of North African natives. The 
Carthaginians were apparently rather tall, with a mean male stature of 
168 cm. The Greek evidence, already quoted, indicates that they were 

There can be no doubt that the majority of the Carthaginians who were 
buried in these tombs were either the descendants of seafarers from Pales- 
tine and Syria, or at least immigrants from the east of similar race. Nine 
skulls of important men, taken from elaborate stone sarcophagi, belong 
to exactly the same type as the majority of the others, except that these 
representatives of the privileged classes had larger heads in all or most 
dimensions than those of the masses. This correlation between size and 
status, or size and opportunity, is a familiar human trait wherever there 
are social and nutritional differences, and has no coincident racial signifi- 
cance. Single Phoenician skulls from two points in the western Mediter- 
ranean, Melilla in the Moroccan Rif, and Ibiza in Spain, 53 conform exactly 
to the standard set by the Carthaginians. 

The last of the three non-Indo-European speaking ethnic groups, the 
Etruscan, probably came to Italy as early as the first quarter of the tenth 

62 Bertholon and Chantre, Rtcherches Anthropologiques dans La Berberie Orientale, pp. 251- 
266. Also: 

Gollignon, R., Anth, vol. 3, 1892, pp. 163-172. 
Mantegazza, P., APA, vol. 6, 1876, pp. 17-29. 
68 Barras de Aragon, F. de las, AMSE, vol. 9, 1930, pp. 35-64; 79-105. 


century B.C. Another wave is said to have arrived in the eighth century. 
The colonists apparently kept up contacts with their homeland until 
about 650 B.C. This homeland, according to the classical tradition, main- 
tained by all Greek and Roman historians from Herodotus to Pliny, was 
Lydia in Asia Minor. That this tradition is accurate is the belief of most 
modern classical scholars. 54 

The cranial evidence from Etruscan tombs 55 substantiates the belief 
that these non-Indo-European, non-Semitic speakers were typical ex- 
amples of the earlier Bronze Age population of the eastern Mediterranean. 
As with the earlier el Argar people of Spain, a mesocephalic mean for 
the cranial index covers the presence of pronounced long heads and round 
heads, with the two extremes, in this case, forming about equal propor- 
tions. Actually, the metrical characteristics of the two series are much 
alike, but the Etruscan skulls were a little larger, which is not surprising, 
for the el Argar crania were for the most part rather small. 

The Etruscan skulls are notably smooth in surface relief, with little in 
the way of browridges; the side walls of the vaults, seen from above, are 
not parallel, as with the longer Mediterranean forms, but converging, 
with the greatest breadth in the parietals and a narrow forehead; the 
orbits are high and rounded, and the nose narrow. The Etruscans, with 
a typically Near Eastern cranial form, resemble both the Cappadocian 
type found in the Hittite period at Alishar, and the planoccipital brachy- 
cephals which appeared in the Bronze Age cemeteries of Cyprus. By 
Roman times these two varieties had blended, to a large extent, into a 
variable mesocephalic form, to which the Phoenicians as well largely be- 

It would be difficult to overemphasize the importance of the migrations 
of eastern Mediterranean peoples by sea to Italy, Spain, and the islands 
between these two peninsulas in protohistoric as well as in prehistoric 
times. Especially in Spain and Italy, large numbers of peoples immi- 
grated, who added, to the basic Mediterranean population of Neolithic 
origin, Near Eastern elements which may still be discerned among Italians 
and Spaniards today. The debt of the Romans to the Etruscans, genet- 
ically as well as culturally, was especially great. 


While the earliest Metal Age culture was being carried westward 
through the Mediterranean by sea, other agencies conveyed it overland 
into central Europe. As before, the main highroad was the Danube Val- 

64 Schachermeyer, Fritz, Etmskische Fruhgeschichte. 
68 Sergi, G., AFA, vol. 41, 1915, pp. 309-313 ff. 


ley, but this time the center of earliest diffusion was not Bohemia, but Hun- 
gary. A series of crania from Bodrogkeresztur in that country M are uni- 
formly dolichocephalic, with the highest individual cranial index, out of 
more than fifty examples, only 76. This is too low for Danubians of the 
usual Neolithic type, and one suspects a movement from the northeast of 
peoples of Corded origin. The common presence of copper battle-axes, red 
ochre, tumulus burials, and other south Russian cultural traits in Copper 
Age sites in Hungary 57 would tend to confirm this deduction. In the west 
Corded people brought the first metal to Switzerland, and in this case 
crania of definitely Corded type are involved. 58 

The inhabitants of Yugoslavia during the Copper Age were, like those 
of Hungary, also uniformly dolichocephalic. 59 Unfortunately, here also 
we have no further information of racial significance. As one approaches 
the mouth of the Danube, however, this dolichocephalic uniformity dis- 
appears. Four skulls from Russe, in Bulgaria, include one male of Corded 
type, a mesocephalic male, and two brachycephalic females.** 

From this evidence, such as it is, we may deduce that the people who 
brought copper into the Danube Valley at the close of the Neolithic period 
came from two centers, southern Russia and the Caucasus, and Anatolia, 
by way of Troy. The chief carriers were the Corded people or some 
others equally dolichocephalic, while brachycephals from Asia Minor 
were of little importance from the racial standpoint. 

While Copper Age civilization was thus spreading westward along the 
Danube and the lands to the north, a countermovement in the form of the 
Bell Beaker invasion travelled eastward from the Rhine to the Danube, 
and as far as Poland and Hungary. The remains of these Bell Beaker peo- 
ple occupy single graves or groups of graves, rather than whole cemeteries; 
they were apparently wandering traders, trafficking in metals, for their 
gold spirals have been found in Danish graves of the corridor- tomb period. 
They were thus in all likelihood rivals of the Battle-Axe people in their 
search for amber. 

It is not known how they went from Spain to central Europe. Sporadic 
finds in France and northern Italy suggest the Rh6ne-Rhine and the 
Brenner Pass routes as alternatives. 61 In neither case is the evidence very 
satisfactory, and neither excludes the other. From the Rhine Valley as a 

M Bartucz, L., MAGW, vol. 57, 1927, pp. 126-130. 

67 Hillebrand, J., AH, vol. 4, 1929, pp. 1-51. 

68 Virchow, R., ZFE, vol. 17, 1885, p. 288. (2 adult female, and I juvenile, skulls 
from Vinelz). 

B iupanifc, N., RA, vol. 29, 1919, p. 28. 

60 Drontschilow, K., Mitt. Arch. Inst. Sofia, 1924, pp. 187-201, quoted by Sailer, K., 
ZFAE, vol. 77, #5/6, 1925, pp. 515-571. 

61 Childe, The Danube in Prehistory, p. 196. 


center, Bell Beaker expeditions moved eastward into Bohemia, Austria, Po- 
land, and Hungary; those who took part in these movements were even- 
tually absorbed into the local populations. The Bell Beaker people who 
remained in the Rhinelands, however, came into intimate contact with the 
Corded people, who had invaded from the east and northeast, and with the 
corridor-tomb megalithic population to the north, whose domain ex- 
tended down into the Netherlands. These three, of which the Bell Beaker 
element formed perhaps the dominant one, amalgamated to form an 
Early Bronze Age cultural unit, the so-called Zoned Beaker people, who 
invaded England and Scotland as the first important carriers of metal. 

The Bell Beaker physical type is known to us from sixty or more skulls 
from scattered burials in Germany, Austria, Poland, Czecho-Slovakia, and 
Hungary. 62 Of these, about one-third are truly brachycephalic, while 
the others are, almost without exception, mesocephals. In the Rhine 
country around Worms, three-fourths or more of the Bell Beaker crania are 
brachycephalic; in Austria, one finds an equally high ratio; but in Bo- 
hemia and Poland the high brachycephaly becomes less frequent, and at 
Tokol in Hungary, in a series of ten crania, four are mesocephalic and 
six are dolichocephalic. 63 

So high is the mesocephalic ratio, and except for Hungary, so infre- 
quent the truly long-headed crania associated with this type, that the 
mesocephals are clearly one branch of the main type, and not the product 
of local mixture with long heads. Morphologically, the mesocephals are 
essentially Bell Beaker. 

The series of skulls from the Rhineland, including nine adult males, is 
the most suitable for comparison (see Appendix I, col. 21). It is identical 
in the cranial index mean with that of Furst's forty-four male Bronze Age 
skulls from Cyprus, which have already been studied, and which have 
been called Dinaric. The Rhenish crania are a little larger in vault dimen- 
sions, and particularly in height; but are almost identical facially. Mor- 
phologically, the two groups are also similar, but the Bell Beaker group is 
more extreme in many ways; the browridges are often heavy, the general 
ruggedness frequently greater. The faces are characteristically narrow, the 
orbits medium to high, the nasal skeleton high and aquiline; the occiput 
frequently flat. The stature for six males reached the high mean of 177 cm. 

2 Bartels, P., PZ, vol. 5, 1912, pp. 67-82. 
Jankowsky, W., AAnz, vol. 8, 1932, pp. 104-115. 
Palliardi, J., WPZ, vol. 6, 1919, pp. 41-56. 
Sailer, K., ZFAE, vol. 77, #5/6, 1925, pp. 515-571. 
Schliz, A., AFA, vol. 35, 1908, pp. 239-267. 

Sedlaczek-Komorowski, L., BAPS, ser. B, vol. 2, 1932, pp. 253-257. 
Stocky, A., and Matiegka, J., AnthPr, vol. 3, #2, 1925, pp. 138-155. 
Trauwitz-Hellwig, J. von, MAGW, vol. 53, 1923, pp. 251-265. 
8 Bartucz, L., MAGW, vol. 57, 1927, p. 128. 


The deviation of the Rhenish Bell Beaker skulls, such as it is, from the 
Aegean and eastern Mediterranean Dinaric form, lies in a Borreby direc- 
tion. It is, therefore, more than likely that the invaders mixed with the 
descendants of the earlier Neolithic brachycephals, whose territory 
stretched along the North Sea coast from southern Sweden to Belgium. 
On the whole, however, at the period represented by the Worms crania, 
the eastern or Dinaric element was the more important. 

The Spanish Bell Beaker problem now stands in a somewhat clearer light 
than before. The Dinaric type, with which the Rhenish Bell beakers are 
associated, is one which entered the western Mediterranean by sea from 
the east, and eventually moved, by some route yet to be determined in 
an accurate manner, to the north, and eventually to central Europe. 
The paucity of brachycephals in Spain may be due to the paucity of re- 
mains of this culture in general. It is still possible, one might add, that 
certain North African elements became involved in the Bell Beaker racial 
type, but such an accretion is unnecessary and hardly likely. 

The Bell Beaker people were probably the first intrusive brachycephals. 
to enter the Austrian Alps, and the mountains of northeastern Bohemia, 
for the push of Lake Dwelling Alpines southeastward toward the Balkans 
happened later in the Bronze Age. It is, therefore, possible that the present 
Dinaric populations of the Dinaric Alps and the Carpathians may be 
derived in part from this eastward invasion. The small numbers and scat- 
tered burial habits of the Bell Beaker people on the more densely popu- 
lated plains of Europe must have made them of much less ethnic impor- 
tance there than in the mountains. 

In their Rhineland center, the more numerous Bell Beaker people had 
constant relationships with the inhabitants of Denmark, who were still 
burying in corridor tombs. Furthermore, the Corded people, one branch 
of whom invaded Jutland and introduced the single-grave type of burial, 
also migrated to the Rhine Valley, and here amalgamated themselves 
with the Bell Beaker people, who were already in process of mixing with 
their Borreby type neighbors. The result of this triple fusion was a great 
expansion, and a population overflow down the Rhine, in the direction 
of Britain. 


The consideration of the Bell Beaker problem leads naturally to that of 
the Bronze Age in the British Isles, where the Beaker people found their 
most important and most lasting home. Coming down the Rhine and out 
into the North Sea, they invaded the whole eastern coast of England and 
of Scotland, and also the shore of the Channel. 

The Beaker invasion of Britain was not a simple affair. Not only did the 


newcomers land in many places, but they brought with them somewhat 
different traditions. Although most of them brought zoned beakers and 
battle axes, in consequence of their blending with the Corded people in 
the Rhinelands, others, with the older type of bell beakers and with stone 
wrist-guards of Spanish inspiration, seem to have entered unaffected by 
Corded influence. 

Like their predecessors the Long Barrow people, the new invaders who 
went to England chose open lands for settlement, and eschewed the forest 
of the Midlands, and the Weald of Surrey, Sussex, and Kent. Yorkshire 
with its moors was a favorite spot, while other centers were Wiltshire and 
Gloucestershire in the south, and Derbyshire and Staffordshire in be- 
tween. 64 On the whole, the Beaker people chose the same regions which 
had attracted the builders of the long barrows, except that the concen- 
tration in Yorkshire was an innovation. The Beaker people did not ex- 
terminate the Long Barrow people, who continued for a while to build 
their characteristic earth-covered vaults, in some of which Beaker pots 
have actually been found. The remains of the newcomers, however, are 
always buried singly under round barrows, of a type which the Corded 
people contributed to the Zoned Beaker complex. 

In comparison with the Continent, Great Britain contains a great plenty 
of Beaker skeletal material. The invasions which reached this island 
brought the wholesale migration of a large population. Over two hundred 
and sixty crania from England alone have been preserved and studied. 
Out of a series of one hundred and fifty exhaustively analyzed by Morant, 
the brachycephals exceed the pure long heads in the ratio of three to one, 
while the intermediate forms are about equal in number to the latter. 
This segregation would indicate that the blending between the Corded 
racial element and its round-headed companions was incomplete at the 
time of invasion, as well as afterward. In all the regions from which a 
considerable number of skulls have been taken, the proportion between 
round heads and long heads is constant, and this would indicate that the 
survivors of the Long Barrow people were not buried in the tombs of the 

The Bronze Age people of England, as represented by this Beaker 
series, were clearly heterogeneous. The three ancestral elements which met 
in the Rhinelands may be distinguished easily. All three were tall, and 
the mean stature of the whole group was about 174 cm. 68 The Corded ele- 
ment, however, was the tallest, and the Borreby element, about 170 cm., 
the shortest. On the whole, the heavy-boned, rugged quality of the 

454 Morant, G. M. Biometrika, vol. 18, 1926, pp. 56-98. 

65 Obtained by applying Pearson's formula to 27 adult male femora listed by Thur- 
man. Thurman, J., MASL, vol. 1, 1865, pp. 120-168, 459-519; vol. 3, 1867, pp. 41-80. 


Borreby type seems to have influenced the bodily build of the total group. 

The Beaker skulls as a whole are large, long, and high vaulted, whatever 
their shape. They form one of the rare groups in the world with a cranial 
length of 184 mm. and an index of over 80. This peculiarity they share 
with the few known brachycephalic crania of the Upper Palaeolithic. 
Again reminiscent of Upper Palaeolithic skulls is the ruggedness of muscu- 
lar markings, the prominence of browridges and occipital lines, and the 
depth and breadth of the mandible. 

In the Crania Britannica are engravings of seventy- three male crania 
of this group; by observing them morphologically it is possible to segregate 
them into their component elements. Twenty-four, or one-third of the 
whole, are planoccipital. This ratio is probably about the correct propor- 
tion of the original Bell Beaker element in the blend, with the Corded 
group one-fourth, and the rest Borreby. The planoccipital skulls are, as 
one would expect, the most brachycephalic; for over sixty per cent of all 
crania over the index point 83 possess some posterior flattening. 

When seriated by index groups and occipital form, the planoccipital 
brachycephalic male crania (see Appendix I, col. 22) approach metrically 
the series already discussed from Worms, as well as that from Bronze Age 
Cyprus. The British planoccipitals are larger vaulted, in all three dimen- 
sions, than their continental and Near Eastern prototypes; they are also 
wider faced; but in total and upper face heights and in nasal dimensions, 
they are much the same. The curvoccipital brachycephalic crania (see 
Appendix I, col. 23) are much larger; and it is this element which contrib- 
utes the combination of a truly long vault with a high index. They like- 
wise have large faces, of great width, and of great mandibular size. One 
of the most striking differences between the two brachycephalic British 
sub-groups lies in the disproportion of face heights. Both have the same 
upper face height; but the total face height, from nasion to men ton, is five 
mm. greater in the curvoccipital group. The lower jaw of the planoccipi- 
tal skulls is more nearly of a normal Dinaric form, while that of the Borreby 
element is nearly equal to Upper Palaeolithic standards. 

The dolichocephalic crania (see Appendix I, col. 24), forming the least 
numerous of the three elements, are of pure Corded type, and furnish an 
opportunity to study this form in greater numbers than elsewhere. The 
vault is very long, and extremely high, with a breadth-height ratio of 105, 
and extremely long faces, with deep, narrow mandibles. There can be no 
question that these most extreme variants from the fundamental Mediter- 
ranean stock came to England as part of the Zoned Beaker racial complex, 
and do not represent accretions of megalithic Long Barrow survivors, 
although both elements, in England as in Scandinavia, entered into the 
ultimate composition of the living population. 


In Scotland the progress of events in the Early Bronze Age was quite 
different from that in England, and more complicated. The Beaker peo- 
ple who arrived on the eastern shore came in part directly from Holland, 
and in part from England. A few may have approached from the west, 
by way of Wales. At the time of the Beaker arrival, or not long after it, 
another group of people, named after the so-called Food Vessels which 
they placed in their tombs, seem to have arisen in the west, or to have 
arrived there from Ireland, where they were also prevalent during the 
Early Bronze Age. These Food Vessel people buried their dead in in- 
dividual cists, as did the Beaker people, but often incinerated, for which 
reason their skeletal remains are relatively rare. The two groups Beaker 
and Food Vessel had close relationships and interchanged material pos- 
sessions and ideas. In many Scottish cist graves, neither type of pottery is 
present, and it is not always possible to tell to which original complex the 
burial belongs. 66 

The short cist skeletons of Scotland have been lumped together regard- 
less of original cultural affiliation, which in many cases may have been im- 
possible to determine. By this means a series of seventy-seven crania has 
been assembled for study. 67 (See Appendix I, col. 25.) In general, the 
Scottish Short Cist people resembled the Beaker invaders of England, but 
were by no means identical with them. The means of the cranial dimen- 
sions are in many cases smaller, and the larger elements in the blend seem 
to be less in evidence. Furthermore, the stature seems to have been shorter, 
with a mean of 165.0 cm. 68 for seventeen males. The group as a whole is 
more purely Beaker in the continental sense, or Dinaric, than is that in 
England; metrically, the Scottish series resembles the non-Borreby brachy- 
cephalic element in the British Beaker population, and also approximates 
the skulls from the Rhineland. In several features, such as a lower vault, it 
comes closer to the Cypriote Bronze Age group than does any wholly 
Beaker series which we have studied. 

The reasons for the difference between the Scottish and English series 
are not difficult to discover. The Borreby element is less prominent in 

68 Childe, V. G , The Prehistory of Scotland, pp. 81-95. 

67 Morant, G. M., and Reid, R. W. } Biometrika, vols. 3-4, 1928. Later publications, 
mostly in the PSAS scries, would swell this number by at least twelve, but would in no 
way alter the conclusions. 

** Calender, J. G., PSAS, vol. 58, 1924, pp. 23-27. 

Callander, J. G., and Low, A., PSAS, vol. 64, 1930, pp. 191-199. 

Craw, J. H., and Low, A., PSAS, vol. 67, 1933, pp. 308-311. 

Edwards, A. J. H., PSAS, vol. 65, 1931, p. 421. 

Edwards, A. J. H., and Low, A., PSAS, vol. 66, 1932, pp. 418-426; vol. 67, 1933, 
pp. 164-176. 

Gordon, J. T., and Waterston, D., PSAS, vol. 67, 1933, pp. 354-361. 

Low, A., PSAS, vol. 67, 1933, pp. 176-186. 

Ritchie, J., and Dow, D. R., PSAS, vol. 69, 1935, pp. 401-415. 


Scotland, and the same is true of the Corded. In fact, three out of four 
dolichocephalic male crania from short cists seem to be of a Megalithic 
type, while only one has the characteristic vault form of the Battle-Axe 
people. Long heads are less frequent here than in England, and the orig- 
inal eastern Mediterranean brachycephalic type is in the majority. Log- 
ically, one would expect that the Food Vessel people belonged to this racial 

It is impossible, however, to determine with any certainty the physical 
type of the Food Vessel people in Scotland, for only four complete skeletons 
have been associated with this pottery form. Three, however, which are 
males, are all brachycephalic and of medium stature, and belong, in the 
totality of their features, to a small Beaker variety, 69 as does the single 
female. Two other individuals, represented only by long bones, were, 
respectively, 166 and 173 cm. tall. Little is to be learned, unfortunately, 
from the members of this small group, except that they were no different 
from the Beaker people who occupied the same type of cist. 

There is, however, one far better way to discover the physical affinities 
of the Food Vessel people, and that is by a study of the Bronze Age remains 
from Ireland. As far as we know from published evidence, the Beaker 
people never went to Ireland at all. The thirty odd known Irish skeletons 
of the Bronze Age, taken from short cists, were associated with food ves- 
sels in most cases, or at least when there is known to have been any pottery. 

The series as a whole 70 (see Appendix I, col. 26) is tall and slender 
boned; the skulls, almost exclusively brachycephalic, are often thin walled; 
the bony relief is rarely as prominent as in the British specimens. Metri- 
cally, the Irish crania are narrower headed and narrower faced than the 
Scottish, and are almost identical with the Adlersburg group in Germany, 
and quite close to the series from Cyprus. Their most notable difference 
from the British group, which confirms their similarity to the skulls from 
Cyprus, is in their narrow facial breadth. In this and in many other ways, 
the Scottish skulls are intermediate between the English and the Irish. 

9 Dow, D. R., PSAS, vol. 69, 1935, pp. 401-415. 

Low, A., PSAS, vol. 64, 1930, pp. 191-195; vol. 65, 1931, pp. 418-426. PAAS, 
1904-06, pp. 133-142. 

Waterston, D., PSAS, vol. 67, 1933, pp. 354-361. 

70 A composite group from the following sources : 

Haddon, A. C., PRIA, vols. 3-4, 1896-98, pp. 570-585. 

Martin, C. P., JSAI, vol. 62, 1932, p. 55; vol. 64, 1934, pp. 87-89. 

Martin, C. P., Price, L., and Mitchell, G. F., PRIA, vol. 63, 1936, sec. C, #7. 

Movius, H. L., PRIA, vol. 61, 1934, pp. 258-284; JSAI, vol. 59, 1929, pp. 99-115; 
vol. 64, 1934, pp. 73-85; vol. 65, 1935, pp. 213-222. 

Shea, S., JGAS, vol. 12, 1925, pp. 13-22. 

See also: 

Martin, C. P., Prehistoric Man in Ireland. 

Morant, G. M., JRAI, vol. 66, 1936, pp. 43-55. 


The Irish Bronze Age people who were buried in association with food 
vessels were, therefore, members of the racial type which was originally 
linked with the Beaker complex, without the associated Borreby and 
Corded elements. Childe finds possible prototypes of the food vessels both 
in Germany and in Spain. 71 Without doubt, in any case, there were move- 
ments from northern Spain and the western end of the Pyrenees during 
the Bronze Age, which brought halberds to Ireland, and thence to Scot- 
land, along with other cultural innovations. These movements were quite 
late, but so, in all probability, was the spread of the Food Vessel people, 
who often incinerated. 

It is necessary to choose between two routes of invasion for the Food 
Vessel people, for they were obviously not indigenous. The first, from 
Germany and Holland, would be somehow separate from the Beaker in- 
vasions, but yet would bring the most basic Beaker physical element. The 
second is from Spain, where the Beaker people were probably only one of 
a number of related brachycephalic groups. The latter seems the more 
likely, purely on racial grounds; furthermore, on the Scottish food vessels 
there are often cord impressions, on the Irish there are none. The direc- 
tion, therefore, was probably from Ireland to Scotland and not vice versa. 


In the Early Bronze Age there were, aside from the Aegean, three im- 
portant cultural centers in Europe southeastern Spain, Britain, and cen- 
tral Europe. We have already dealt with the first two and studied the 
racial derivations of their peoples. In central Europe, the center of civili- 
zation was again on the Danube; in Bohemia, Moravia, Silesia, Lower 
Austria, and Saxo-Thuringia. The Bronze Age culture of this Danubian 
region is called Aunjetitz (Unetice) after an important site in Bohemia. 

The origins of this Aunjetitz culture were multiple. The elements of 
which it was composed include: the basic local Neolithic and Copper Age; 
northern influences which were mostly Corded; the Bell Beaker invasion; 
and metallurgy from Anatolia and the Aegean, coming directly over- 
land. 72 

The evidence as to the racial composition of this culturally heteroge- 
neous population is fortunately abundant and clear. A number of large 
and well-analyzed series makes it possible to determine its nature without 
much doubt. On the whole, the group is moderately varied. Three major 
elements are involved: the short, moderately dolichocephalic, high- 
vaulted, small-faced Danubian Neolithic type; the?' familiar Corded form, 
and some brachycephals, in moderate numbers, which are probably for 

n Childe, The Prehistory of Scotland, pp. 89-95. 
72 Childe, The Bronze Age, pp. 139-140. 


the most part of Bell Beaker origin, although the same racial type may 
have come up the Danube from the Black Sea and the Aegean. Dinaric in- 
fluence is most evident in the earliest Aunjetitz sites of Lower Austria, as at 
Hainburg-Teichtal, 73 but it disappeared shortly through absorption. 

One of the most fruitful groups for examination is that of the skeletons 
from the Lower Austrian cemetery of Gemeinlebarn. 74 Here fifty-one 
adult crania were found which were in condition for study, to which have 
been added twenty-five others from smaller sites in Lower Austria and 
Moravia, making a total of forty-seven male and fifty- two female skulls, 
as well as a large number of associated long bones. 

The mean stature of the males is 165 cm., a moderate figure, lying 
between that of the earlier Neolithic Danubians and the Corded people, 
as represented in the larger series in which the latter appear, in Scandi- 
navia and England. The limb proportions show a greater length of the 
distal segments in both arms and legs than is the case with most historic 
Germanic or Nordic skeletons the Lower Austrian Aunjetitz people 
resembled their Neolithic ancestors in this respect. The bones, however, 
are quite heavy and powerful, and show that they must have had wide 
and heavy shoulders. 

The crania (see Appendix I, col. 27) belong metrically about half-way 
between the Corded and Danubian Neolithic means in almost every 
character; the only exception being a slight addition in the head breadth 
dimension which might be attributed to the inclusion of a few brachy- 
cephals. The cranial index, which varies individually from 64 to 85, 
centers about a mean of 74; and dolichocephaly is the prevailing form. 
The profile of the skull as seen from above usually takes one of two forms 
a long oval with almost parallel sides, which is the Corded type, and a 
pentagonoid, or "shield shape," which is the Neolithic Danubian. The 
vaults are high, in most cases higher than the breadth, a feature which is 
derived from both of the principal ancestral types. The face is quite long 
in both segments, and narrow. Although, the mean nasal index is mesor- 
rhine, a little less than half of the series is leptorrhine. The orbits, like 
those of both earlier strains, are of moderate height. 

In the male series, only four crania have indices over 78, and all of these 
are curvoccipital. One of them, with heavy browridges, a wide interorbital 
distance, a wide, deep jaw with everted gonial angles, and no canine fossa, 
looks like some intruder from the northern European forests, such as we 
have already met in the Neolithic; while another, which is hyperbr achy- 
cephalic, has an extremely narrow face and jaw, and may be either an 
Anatolian or a Beaker remnant. 

78 Geyer, E., MAGW, vol. 60, 1930, pp. 65-140. 
w Szombathy, J., MAGW, vol. 64, 1934, pp. 1-101. 


The group as a whole has a normal to excessive development of the 
browridges, and a narrow-rooted form of the nasal bbnes, which spring 
prominently from a considerable nasion depression. The continuous 
fron to-nasal profile of Near Eastern Bronze Age skulls is apparently alien to 
this composite type. 

The above discussion could be applied without much change to other 
Austrian Aunjetitz series, notably that from Stillfried, 76 which includes 
nine males. Here the ratio of factors involved may be slightly differ- 
ent, for the cranial index mean is mesocephalic, and the nasal index 
purely leptorrhine however, the group contains no brachycephals. 

Nearly a hundred crania from Bohemia, collected from a number of 
sites 76 (see Appendix I, col. 28), are on the whole extremely dolichocephalic, 
with a mean index of 71. A series of thirty-two males 77 (see Appendix I, 
col. 29), like the Austrian group, is again intermediate in most if not all 
measurements between the Corded and Danubian Neolithic means. As 
with the Gemeinlebarn series, the longest crania are the highest, and 
possess the longest faces. A Corded-Danubian cross, with a very little 
Dinaric (since the highest indices go up to 83) is indicated. This hybrid 
form, as will be seen later, may be given the name "Nordic" in the skeletal 
sense, since it seems identical with that of historic Nordic peoples living 
in the same area. 

The stature of Bohemian and Moravian Aunjetitz males, as with those 
from Lower Austria, is about 167 cm. 78 This is considerably less than the 
Corded stature for Scandinavia, and that of the British Bell Beaker long 
heads, but more than that recorded in the central European Corded series 
of Neolithic date. Either our groups are too small for accuracy, which is 
quite probable, or else the Corded people of central Europe were not as 
tall as those who invaded the far northwest. At any rate, the Aunjetitz 
people of central Europe are less exaggerated in head and face dimensions 
than those whom we have previously studied, and anticipate the " Nordic" 
peoples of the Iron Age. 

Around the peripheries of the Upper Danubian center, modifications of 
the standard Aunjetitz racial amalgam occurred. In Saxony and Thu- 
ringia, where there was an especially strong Corded cultural element, the 
coincident type was of course equally strong. 79 But on the Rhine, the Bell 
Beaker -cultural influence continued, and brachycephals also persisted. 80 
In the Tyrol and Upper Austria, Binaries of the Bell Beaker type re- 

Schurer von Waldheim, Hella, MAGW, vol. 39, 1919 f pp. 247-263. 

TC Stockf, A., AnthPr, vol. 9, 1931, pp. 225-275. 

77 Hellich, B., Praehistorickl lebky v &cMch & Sbirky Musea Kralovstvi Ceskeho. 

7 Matiegka, J., MAGW, vol. 41, 1911, pp. 348-387. 

TO Hcberer, G., VGPA, vol. 8, 1937, pp. 59-68, 

As at Rheinsheim. Easier, A., MAGW, vol. 55, 1925, pp. 261-266. 


mained firmly ensconced, 81 where their survival in this mountain refuge 
was destined to be permanent. 

About forty skulls are known from the Bronze Age sites of Switzerland. 82 
The most important fact to be deduced from them is that the old Neolithic 
elements persisted with little change. An infiltration of Aunjetitz culture 
was accompanied by the addition of some Corded types to the group, and 
in the meanwhile a few planoccipital brachycephals of Bell Beaker type ap- 
peared. On the whole, the Swiss seem to have become slightly longer 
headed during this period, probably due in large part to Aunjetitz in- 

It is impossible to carry this survey of the Early and Middle Bronze Age 
racial types in central Europe much farther to the westward. We have 
already seen that brachycephals of the type which spread through the 
Mediterranean during the Bronze Age entered the southern departments 
of France, near the eastern end of the Pyrenees and the Gulf of Lyons 
(page 149). Aside from this Spanish overflow in the south, the French 
Bronze Age was largely confined to two other peripheral points-r-Savoie 
and Franche Comt6 and Brittany in the extreme west. 

On the northeastern flank of France, in Franche Comt6, a number 
of skeletons have been taken from tumuli which apparently date from 
the Middle Bronze Age, a time at which invasions spread over the 
upper Rhine and Jura from the Bavarian highlands into northeastern 
Gaul. 88 

Seven out of eight skeletons of this period were those of tall, planoccip- 
ital, brachycephals, 84 who belonged, as far as one can tell, to a Bell Beaker 
type familiar in earlier times in the Rhinelands. Two tumuli of later 
date contained high-vaulted dolichocephalic crania, belonging to small- 
statured individuals, like the single dolichocephalic example from the 
earlier group. Thus, as far as we can tell, a Bell Beaker type, associated 
with an older Danubian Neolithic element, entered northeastern France 
in the Middle Bronze Age from the highland belt of southern Germany, 
south of the central Aunjetitz range. 

In Brittany, the earliest metal industry was mostly of the Middle Bronze 
Age; round barrows were built apart from the megalithic tombs, which 
were still used by the descendants of the bringers of that cult to the Atlantic 
seaboard. In one cemetery, that of Saint-Urnel en Plomeur in Finist&re, 

si Meyer, A. B,, MAGW, vol. 15, 1885, pp. 99-106. 

82 Pittard, E., Anth, vol. 10, 1899, pp. 281-289; vol. 17, 1906, pp. 547-557; ASAG, 
vol. 7, #1, 1934, pp. 1-7; RA, vol. 45, 1935, pp. 5-12. 

Schenk, A., BMSA, ser, 5, vol. 8, 1907, pp. 218-228; REAP, vol. 15, 1905, pp. 389-407. 
Schlaginhaufen, O., BSGA, vol. 2, 1926, pp. 15-24; MAGZ, vol. 29, 1924, pp. 220-241. 
w Childe, The Bron& Age, p. 174. 
84 Piroutet, M., Anth, vol. 38, 1928, pp. 51-60. 


tall, dolichocephalic people with large heads, narrow noses, and robust 
jaws were buried throughout the Bronze Age. 85 There were Beaker people 
in Brittany as well, and one may suppose the presence, in addition, of the 
usual Beaker physical type. 

Aside from these instances there are no Bronze Age remains from 
France which give us a definite picture of the population of any specific 
part of the country.. France, for the most part, failed to participate in the 
great cultural movements of the Bronze Age, and was a backwater in 
which Neolithic and even Mesolithic peoples survived with little change 
in their manner of living. 


During the Early Bronze Age, Scandinavia and the eastern Baltic coun- 
tries had been unable to obtain enough metal for tools and weapons, and 
hence had enjoyed the Late Neolithic efflorescence which we have already 
studied. Their first real metal period, therefore, was the Middle Bronze 
Age, later than the first Beaker settlement in England, or the Aunjetitz 
development in central Europe. 

The Scandinavian Bronze Age probably began about 1500 B.C., and 
lasted for nearly a thousand years. It was a period of great prosperity, for 
Jutish amber brought bronze and gold objects to the north in trade. The 
limits of this cultural center, however, were restricted. Most bronze has 
been found in Denmark, since in Sweden and southern Norway metal 
was dear, and seldom discarded in graves. North of the sixty-eighth 
parallel of north latitude, the Arctic stone age prevailed throughout this 
period on the coasts of the Arctic Ocean and in the forests and mountains 86 
of Norway and northern Sweden, as well as in Finland. 

During the Middle Bronze Age, cremation, which had begun elsewhere 
as early as Danubian Neolithic times, gradually crept in as a major sub- 
stitute for the earlier inhumation, and by the beginning of the Late Bronze 
Age, it had become the only method of disposing of the dead. For this 
reason skeletal material from the five hundred year stretch of the Middle 
Bronze Age becomes progressively scarce. 

In Sweden we are limited to some twenty-one skulls, of which thirteen 
are those of males. 87 They belong to types already familiar to us from the 
Neolithic, and show no change of population. If anything, however, the 
long-headed elements are even more in evidence, and the head form is 
prevailingly dolichocephalic. In Denmark again, -twenty seems to be the 

86 Le Pontois, Bernard, Le Finistire prSkistorique. 
88 Shetelig, Falk, and Gordon, pp; 170-172. 
w Arbo, G., FVO, 1901. 
Hillebrand, B. E. f ATS, 1864. 
Retzius, G., Crania Suecica; Ymcr, 1900. 


limit; 88 and here the old Neolithic population survived without per- 
ceptible alteration. The Bronze Age men were as tall as their predecessors, 
with a mean stature of 172 cm.; and the blend of long- and round-headed 
types struck the same high mesocephalic mean. 

There is evidence that some of the Danes of this period were blond, since 
the hair, teeth, and clothing of a young woman, buried at Egtved, Jut- 
land, were perfectly preserved by the tannic acid from the oak coffin in 
which she lay, under a mound. This hair, cut short on the forehead and 
hanging in a long bob at the rear, was apparently straight as well as fair. 
Unfortunately, the bones were not also preserved, and it is impossible to 
tell to which of the prevalent Neolithic and Bronze Age Danish racial 
types she belonged. 89 

On the whole we may be reasonably confident that the Middle Bronze 
Age in Scandinavia involved no important racial change. The same blend 
of at least three peoples, who had combined to create a brilliant Late 
Neolithic, were carried over into the age of metal. 

In the far north of the Scandinavian Peninsula, out of reach of all but 
the most remote Bronze Age influences, we are led, on archaeological 
grounds, to believe that the older peoples continued to lead their simple 
existence. Although there is as yet no direct skeletal evidence of their sur- 
vival, a body of collateral evidence from across the Baltic makes this, by 
parallel inference, certain. 

At various points near the Esthonian coast of the Gulf of Finland, a re- 
markable group of skeletons has been found in cists under tumuli, prob- 
ably dating from about 1200 B.C., near the beginning of the Middle Bronze 
Age, although they may possibly have been as much as seven hundred 
years later. 90 (See Appendix I, col. 30.) Ten male and five female skulls 
belong to one homogeneous racial type, extremely dolichocephalic, with 
a mean cranial length of 195 mm. The faces are very long, and also 
wide; the nose is of great height. The browridges are in many cases heavy, 
and the nasal bones high and projecting, but deep-set under a strong 
glabella. These skulls are similar in many respects to the Corded racial 
type, especially as exemplified by the dolichocephalic element in the 
British Bronze Age population. Like the latter, they are associated with 
long bones which indicate tall stature. The males, in fact, averaged 172 
cm.; the females 165. 

Unlike the Corded group, however, these Esthonian skulls are as large 
in vault and face size as the Upper Palaeolithic group from central Europe, 

88 Nielsen, H. A., ANOH, II, vol. 21, 1906; III, vol. 5, 1915, pp. 360-365. 

Virchow, R. } AFA, vol. 4, 1870, p. 55. 

Coutil, L., BSPF, vol. 27, 1930, pp. 187-189. 

M Friedenthal, A., ZFE, vol. 63, 1931, pp. 1-39. 


and equal the latter in a number of telltale dimensions, including cranial 
length, orbital width, and bizygomatic diameter. In the height dimen- 
sions of the vault and face, the Esthonian crania exceed all known Euro- 
pean groups of any age. 

This is a clear case of the blending of Upper Palaeolithic survivors, who 
had preserved a hunting life in their northern forest, with Corded horse- 
men and cultivators who had penetrated their fastness, bringing them 
their first direct contact with food-producing civilization. If the Upper 
Palaeolithic group survived in Esthonia, it could have done so in Norway 
as well. It is worth noting the exaggeration of the Corded facial and 
cranial heights in the Esthonian mixture, along with the Upper Pa- 
laeolithic retention of gross vault size and of face breadth. This will later 
be encountered in several living North European populations. 


The remaining portions of Europe, for which there is skeletal docu- 
mentation of Bronze Age date, may be studied as a single unit. This con- 
sists of the grassy plain which extends from northern Germany and the 
Baltic states, south of the forests, across Poland and southern Russia into 
Siberia. It must be remembered that during the Bronze Age this plain 
was drier than at present, and that the agriculture of the Neolithic farmers 
had been discouraged to a large extent both by drought and by the in- 
cursion of Battle-Axe people who had first appeared in the Late Neolithic 
in central and western Europe. 

The evidence from Poland, although meager 91 shows that the Corded 
concentration which had taken place some centuries earlier on Polish 
soil had yielded to the smaller dolichocephalic blend already observed 
in Austria and Bohemia. During the earlier Bronze Age, there had been 
a number of Bell Beaker settlers in Poland as well, who may also have left 
descendants. 92 

The Bronze Age Ukrainians, again, belonged to the same "Nordic" 
type, with a mean cranial index of 74, 93 without the excessive vault height 
of the Austrian and Bohemian groups. In Russia the height is less than 
the breadth in most instances. 

In the parts of southern Russia immediately north of the Black Sea, 
from the Kiev government eastward, Bronze Age remains have not been 
studied in a manner sufficient to permit the formation of adequate conclu- 

91 Nine Bronze Age crania have been published by: 
Czortkower, S., AiUhPr, 1932, pp. 212-218. 
Stojanowski, K., PAr, vol. 3, 1925-27, pp. 52-53. 
Tur, Jan, Swiatowit, vol. 3, 1901, pp. 85-93. 

92 Sediaezek-Komorowski, L., BAPS, ser. B, vol. 2, 1932, pp. 253-257. 
M Debetz, G., AntrM, vol. 4, 1930, pp. 43-105. 


sions. What information is available shows that the population was pre- 
sumably long headed and of tall stature. 94 The same is true of the popula- 
tion of the northern slopes of the Caucasus, where the crania are for the 
most part characterized by exceptionally long faces, narrow noses, and 
vaults of considerable height, like the Corded crania farther west, 95 
although some Megalithic forms may also have been present. Some of the 
Caucasian crania, however, are those of small dolichocephals, and a few, 
for the most part females, are brachycephalic. In the latter part of the 
Bronze Age, the people on the northern slopes of the Caucasus practiced 
cranial deformation of the Hittite variety, which reached, in its southward 
diffusion, to Egypt. 

To the east of European Russia, in western Turkestan and south- 
ern Siberia, there was a nucleus of Bronze Age civilization, which had 
cultural connections with the Danube, the Caucasus, Iran, and China. 96 
That the participants in this Bronze Age were men of European racial 
type is very apparent from the remarkable series of one hundred and 
fifteen adult crania from kurgans in the Minussinsk district of southern 
Siberia 97 (see Appendix I, col. 31), near the headwaters of the Ye- 

This country, which is now the home of nomadic tribes of Kirghiz and 
Kalmucks, was, as early as the second millennium before Christ, occupied 
by a population of purely European character. The series, coming mostly 
from the first millennium B.C., while reasonably homogeneous, shows as 
much variability as do most modern groups. The range of the cranial 
index includes all head forms, among which are a few planoccipital brachy- 
cephals, but the mean is dolichocephalic; similarly the faces are prevail- 
ingly long, the noses narrow. In general, although individual crania 
are as large and as long as the most extreme Corded form, the vaults are 
of moderate size, and the height is considerably less than the breadth. 

In lowness of vault and breadth of face, the Minussinsk skulls resemble 
the Ukrainian Bronze Age group. On the whole, they form a far eastward 
wing of the typical Bronze Age population which reached from Austria 
and Bohemia to central Asia and the term "Nordic," in the skeletal 
sense, is as applicable in the east as in the west. One must expect regional 
differences in a racial type covering such an extensive area. In this case 

94 Gochkevitch, quoted by Tallgren, A. M., ESA, vol. 2, 1926. 

Rau, P., ESA, vol. 4, 1929, pp. 41-57. 

8 Broca, P., BSAP, ser. 2, vol. 8, 1873, pp. 572-578. 

Chantre, E., RDAP, ser. 2, vol. 4, 1881, pp. 247-254. 

Sinirnov, M., BSAP, vol. 12, 1877, pp. 541-553. 

Virchow, R., ZFE, vol. 22, 1890, pp. 412 ff. 

M Tallgren, A, M., ESA, vol. 2, 1926. 

97 Goroshchenko, K., Kurgannie cherepa Minusinskago Okruga, OMM, #2, 1900. 

Debetz, G. F., AZM, #2, 1932, pp. 26-48. 


the difference is simply that the vaults are higher and the faces narrower in 
the west, as far as Poland, and the reverse from the Ukraine on eastward. 

The Andronovo or Minussinsk Kurgan culture lasted from about 
1000 B.C. to 1 A.D., and was followed by other cultures, which lasted until 
the eighth century, when the Kirghiz came in. 98 

These later peoples introduced iron, and the habit of making plaster 
death masks. Not only do these masks represent in many cases a long- 
headed, narrow- nosed and often aquiline, and narrow-faced people, but 
the plaster contains in some instances blond hairs pulled out of the beard. 
The head hair, often preserved on the corpses, is usually brown. 

During the fourth century A.D., the physical type definitely changed, 
as one can tell from the masks the face is now wide and flat, the nose 
broad and flat, with a very low bridge. Eye slits are painted blue and 
the hair blue with black lines." Thus not until after the time of the Huns 
were the Nordics of southwestern Siberia replaced by mongoloids. 


The two or three centuries immediately preceding the 1000 B.C. mark in 
central Europe, and a little later in more backward parts, witnessed 
several cultural innovations which mark the beginning of the Late or Final 
Bronze Age. To the physical anthropologist, the most important of these 
was cremation, on account of which our knowledge of race during this 
most important period is nearly at the zero point. This hiatus is especially 
unfortunate, since the findings of the archaeologists make it clear that 
the Late Bronze Age was a time of considerable shifting and expansion of 

In most of Europe, the Sub-Boreal climate gave way to the Sub- Atlantic, 
which brought an increase in cold and dampness, and fostered the growth 
of forest on former grasslands. The area of soil suitable for cultivation grew 
smaller, while the number of people had increased; these factors alone 
were enough to cause displacements of population. Across the plains of 
Asia as well as of Europe, large movements took place; the migrations of 
the Aryan ancestors into northern India through Afghanistan, and into 
the Iranian plateau, were Late Bronze Age phenomena. 

Cremation had begun in Europe, as an alternate funeral rite, early in 
the Bronze Age, and had gradually increased in popularity in the plains 
north of the Alpine mountain barrier. Its chief center of expansion seems 
to have been the central and eastern grasslands, from eastern Germany 
over to Russia, where it was particularly useful fdt nomadic peoples faced 
with the problem of disposing of their dead on frozen ground. 

98 Golomshtok, E., AA, vol. 35, 1933, pp. 319-322. 
w Golomshtok ; E., BUMP, vol. 2, #4, 1933, pp. 40-45. 


The vehicles which diffused this trait over most of Europe during the 
Late Bronze Age are called Urnfields cultures, which arose on the plain 
north of the Carpathians, from Silesia to the Ukraine. From this center 
they spread in all directions. Some went southward over the Alps to Italy, 
while cremation was introduced into Greece before the time of the Trojan 
war. From a secondary center of expansion in the Alpine highlands, a 
special Urnfields diffusion entered the British Isles as a major invasion. 

For obvious reasons, the skeletal remains associated with the Urnfields 
cemeteries may be disposed of very briefly. Cremated bones which have 
survived the rite are usually so fragile that little in the way of racial 
identification has been attempted, although it has been shown by ex- 
periment that they shrink little or none in the fire. 100 Those from the 
British Isles indicate in general that the invaders of this time may have 
been smaller and slighter than their predecessors. A small series of crania 
from southern England which escaped cremation were those of Alpines 
of the brachycephalic Lake Dwelling type, 101 brought from the secondary 
Urnfields center in Switzerland. On the other hand, eight Late Bronze Age 
skulls from northwestern France 102 are all meso- or dolichocephalic; and 
may have come directly from Germany with the vanguard of the Keltic 
migrations. Eight other skulls, from the Ukrainian urnfields, 103 are long 
headed, and similar to the immediately preceding "Nordic" type of the 
same region. 

Some of the south Russian and Caucasian remains already studied are 
of Late Bronze Age date, as are those from Siberia, both having escaped 
cremation. The general time scale of cultural phenomena in central 
Asia as compared with Europe would indicate that important ethnic move- 
ments were not passing from east to west at that time. By the end of the 
Middle Bronze Age, the ethnic elements which were to form the popula- 
tion of Europe at the beginning of the Iron Age had all arrived; during the 
period of cremation, no new ingredients were added, but those already 
there participated in a considerable readjustment and recombination. 


The Bronze Age covered, in most of Europe, the brief span of some six 
centuries, as compared with an expanse three times as long in Egypt and 
Mesopotamia. During these six centuries, however, important racial 
changes took place in many parts of the European world, while in the 
two valleys from which European civilization emanated, the personnel 

100 Movius, H. L., Jr., PRIA, vol. 61, 1934, pp. 282-283. 

101 Keith, Sir A., JA, vol. 11, 1931, pp. 410-418. 

102 Bouchet, Dr., Anth, vol. 16, 1905, pp. 309-316. 
Piroutet, M., Anth, vol. 38, 1928, pp. 51-60. 

i Debetz, G., AntrK, vol. 4, 1930, pp. 93-105. 


remained constant. The parts of Europe most affected by Bronze Age 
movements of people were the north and west; and hence these activities 
may be interpreted as a late phase of the displacements initiated by the 
retreat of the last glacier, and continued by the discovery of the principles 
of food production. By the end of the Bronze Age, the centers of civiliza- 
tion had begun their movement northward and westward, toward Greece 
and Italy, movements which were later to push much farther in the same 
direction. It is perhaps no coincidence that, since the beginning of the 
Neolithic, people from the east and south had migrated to the north and 
west ahead of this progression. 

Among the problems left over from the Neolithic which the evidence 
of the Bronze Age has helped to clarify is that of the immediate origin of 
the Danubians. In the Neolithic Danubian-like peoples cultivated the 
rich soil of southern Russia and of western Turkestan. We now know that 
they must have formed a large bloc of agriculturalists occupying Asia 
Minor as well, and probably also the Caucasus. Thus they may have 
come into the Danube Valley from either southern Russia or Anatolia, 
or both; and their earlier derivation from the agricultural highlands is 

A second problem, which arose only during the Bronze Age, is the origin 
of the new racial type which appeared, shortly before 2000 B.C., apparently 
from nowhere, in Asia Minor, Palestine, and Cyprus. This new type was 
tall, round headed and frequently planoccipital; its nose was prominent 
and narrow; its face triangular and of moderate length. In its associated 
morphological features, it forecast the appearance of the Dinaric race. 

Brachycephals of this type followed the old Megalithic sea route to 
Italy, the Italian islands, and Spain. In Spain some of them seem to have 
associated themselves with cultural phenomena known as the Bell Beaker 
complex. As the Bell Beaker people, these newcomers travelled from Spain 
to the Rhinelands and to central Europe, where they were the first dis- 
seminators of metal. Having appeared in the Rhineland in considerable 
numbers, they mixed with the older Borreby sub-stratum which had re- 
mained there since the Mesolithic, and with Corded people coming from 
the east. This triple combination moved bodily down the Rhine and 
across the North Sea to Britain. Thus, during the Early Bronze Age, 
England and Scotland were invaded by people of entirely new types, 
who came in numbers sufficient to change the population of these coun- 
tries in a radical manner. At the same time, other movements of these 
brachycephals from the eastern Mediterranean passed by sea from Spain 
to Ireland and from Ireland across to Scotland. 

The appearance of these early Binaries on the Asiatic and European 
scene marks the advent of the third important brachycephalic racial type 


which we have encountered in our survey of the post-glacial prehistory of 
the white race. Unlike the Borreby and Alpine types, it cannot be easily 
or plausibly explained as a simple Palaeolithic survivor. Facially it is 
basically Mediterranean; it seems to be a Mediterranean type brachy- 
cephalized by some non-Mediterranean agency. 104 

These Binaries did not come from central Asia, nor from Mesopotamia 
or Egypt. Facially, they resemble the dolichocephalic residents of Asia 
Minor and the eastern Mediterranean coast lands of the period during 
which they first appeared, in that both have in common a high-bridged, 
high-rooted nose, high orbits, and a sloping forehead. Until further evi- 
dence is found, it is safer to hold that the culture-bearing Binaries of the 
Bronze Age developed in the Syrian highlands, where a similar type of 
brachycephaly is now present, than to try to bring them from a distance. 

Another Bronze Age event of racial moment was the gradual disappear- 
ance through amalgamation of the Corded people and of the Banubians, 
and the emergence of an intermediate long-headed form. This latter, 
which inhabited the immense stretch of territory from Germany and 
Austria to the Altai Mountains, occupied an intermediate position in the 
total roster of greater Mediterranean racial variations. 

In Austria and Bohemia the high vault and narrow face of both Corded 
and Banubian strains persisted, but from southern Russia over to the 
Altai, the vaults were lower and the faces broader. Two variants thus 
appeared, a western and an eastern. There is evidence that the eastern 
group, at least, was partly if not prevailingly blond. Both eastern and 
western divisions may with some confidence be compared to the "Nordic" 
peoples who appeared historically during the Iron Age. 

At the end of the Bronze Age, for a period of two or three centuries, the 
pall of cremation falls over the racial history of Europe. When the smoke 
has lifted during the Early Iron Age, we shall see what changes have taken 
place during this period of darkness. 

104 The principle of Dinaricization will be explained in Chapter VIII, section 6, 
and Chapter XII, sections 11, 12, and 17. See also legend, Plate 35. 

Chapter VI 


In the preceding chapters, we have found it necessary to use archaeol- 
ogy as a system of landmarks by which to chart the movements of human 
groups and their relationships with one another; this study of race in 
terms of culture was essential. Ideas are originated, diffused, and con- 
served by people, and people interbreed. A complete and sudden re- 
placement of one culture by another implies a drastic change of per- 
sonnel, while a gradual merging of a new culture with an old one must 
equally imply the survival, at least in part, of the older population. By 
following these rules we have seen that racial and cultural movements 
are truly connected, and in no instance in which the skeletal record is 
adequate could any contradiction be seen. 

The subject of this book, however, is race, not culture; although culture 
in the archaeological sense has been a valuable guide. But once we arrive 
at the period of history it is no longer necessary to deal exclusively with 
pots and axes and methods of burial; we may consider people as linguistic 
and political groups, with known names and ethnic relationships. This 
has already been possible with the civilized nations of preclassical antiq- 
uity, such as the Egyptians, the Sumerians, the Babylonians, and to a cer- 
tain extent with the Cretans and Hittites, whose writings have so far fur- 
nished little or nothing in the way of documentary information, as well as 
with the early ancestors of the Greeks. 

The peoples of central and northern Europe did not learn to write 
until relatively recent times in most instances well after the beginning of 
the Christian era, and in some cases only within the current millennium. 
But their identities are in many instances known to us from the writings 
of the classical geographers and historians, and, in the Dark Ages, from 
Arabic sources as well. Farther east, in central Asia, the diligence of 
Chinese historians has been of great assistance. In our study of the early 
part of the Iron Age, archaeology will still be needed; but by the time of 
the Christian era it will be possible, for our purposes, to dispense with it 
almost completely, for in treating fully historical and living cultures, 
language serves as the best-known, most easily designated, and most con- 
venient framework available for the creation of units suitable for racial 



Heretofore, we have said little about language. The speech of the 
peoples with whom we have dealt has been unknown to us in almost all 
instances. The exceptions are few: The Egyptians, as we well know, spoke 
a language of the Hamitic stock, with considerable Semitic influence. The 
Babylonians and Assyrians spoke Semitic, while the Sumerian language, 
although it can be read, has not yet been related with certainty to any 
other known tongue or linguistic family. 1 During the third millennium, 
therefore, Hamitic and Semitic languages were used by civilized peoples, 
as was the still unclassified Sumerian. 

Besides these known linguistic groupings found in antiquity, there was 
another group or rather collection of languages spoken in the eastern 
Mediterranean and Asia Minor. These included Lydian, and its probable 
derivative Etruscan; languages of the Caucasus, some of which still sur- 
vive; a few languages of the Himalayas, such as Burushaski; 2 and a whole 
group in Greece and the Aegean Islands, if not farther west, known to us 
almost entirely by place names. Cretan may possibly have also belonged 
to this class of languages. 

A school of linguistic experts headed by the late Professor Marr, and 
championed in the English-speaking world by Dr. Ephraim Speiser, 3 
would group all of these languages together, including a whole row of ex- 
tinct tongues stretching around the so-called "Fertile Crescent" from 
Syria to Elam. The name given this group is "Japhetic," coined to com- 
plete, with Hamitic and Semitic, a Biblical trinity. The living examples 
of this alleged class or family of languages, notably Georgian and Circas- 
sian, employ a number of sounds unfamiliar to the Indo-European, Se- 
mitic, and Hamitic families, and reminiscent of American Indian languages. 

No one denies the wide distribution and importance of these languages 
in ancient times, but there is serious doubt that they may be united into 
a single stock comparable to Semitic, Hamitic, Indo-European, etc. It is 
more likely that this grouping includes a number of independent families, 
but at present it is too early to say what these may be; especially since 
most of them are extinct and will never, in all likelihood, be resuscitated. 
At any rate, it is probable that some of the seafarers of the Late Neolithic 
and of the Bronze Age who migrated westward along the Mediterranean 

1 The supposed kinship between Sumerian and Finno-Ugrian cannot easily be eval- 
uated, owing largely to the gap of over three millennia between the known forms of each. 
Both groups are agglutinative, but the grammatical structure of Sumerian also has ver- 
bal prefixes, often with personal tone, unknown in modern Finnic or Ugric. Sumerian, 
like modern Finnic, Ugric, and Turkish, seems to have vowel harmony. In vocabulary 
there are few similarities. On the whole, this relationship cannot at the moment be 
proved or disproved. Personal communication from Dr. J. Dyneley Prince. See also 
the Prolegomena of his Materials for a Sumerian Lexicon. 

2 Lorimer, D. L., The Burushaski Language. 
8 Speiser, E., Mesopotamian Origins. 



B m e a Borreby \ 
Mordic oooo Alpine 

+KJWW*,. . ,..,. A^AAQjnaric 



.*. -Northern / 
. ; Forest people/ 





to Italy, the Italian islands, and Spain, and thence to Britain, France, 
and Scandinavia, spoke languages derived from the eastern Mediter- 
ranean. It is furthermore possible that modern Basque may be the only 
survivor of this linguistic migration; but this suggested relationship, 
referred to in the preceding chapter, must by no means be accepted as a 

We do not know the languages of the Early Neolithic swineherds who 
introduced a food-producing economy to Spain and western Europe, in- 
cluding the lake shores of Switzerland, and we are not likely to find out. 
We do not, furthermore, know what medium the Danubians who per- 
formed the same pioneering function in another quarter used. The 
speech of the Corded people is equally unknown, and the oldldioms of the 
Palaeolithic survivors in the far north, of the midden dwellers of Denmark, 
and of the Azilian survivors in Switzerland, are far past reconstruction. In 
Europe we must start as late as the Iron Age in our attempt to allocate 
languages to cultural or racial groups. 

Today the members of the white race speak languages of the following 
linguistic stocks: Semitic, Hamitic, Indo-European, Ural-Altaic, 4 Euska- 
rian (Basque), and various languages of the Caucasus and Himalayas, 
which it would be futile to attempt to classify here. At present the two 
most important are Indo-European and Ural-Altaic. Yet in antiquity, 
while civilization of the first water was in the hands of Hamites, Semites, 
and Sumerians, all Indo-European and probably most Ural-Altaic speak- 
ers, if they existed as such, were illiterate barbarians. 

Indo-European languages are spoken by more white people today than 
are all of the others put together, several times over. People speaking 
Indo-European languages have monopolized the cultural advances of 
modern science; but it must not be forgotten that, as late as the Middle 
Ages, Semites, Turks, and Chinese were more advanced than the majority 
of Indo-European speakers. The linguists tell us that the Indo-European 
speakers did not initially domesticate one useful animal, or one cultivated 

Linguistically, Indo-European is probably a relatively recent phenom- 
enon, which arose after animals had been tamed and plants cultivated. 
The latest researches find it to be a derivative of an initially mixed lan- 
guage, whose principal elements were Uralic, called element A, and some 
undesignated element B which was probably one of the eastern Mediter- 
ranean or Caucasic languages. 5 The plants and animals on which the 

4 Concerning the question of Ural-Altaic unity, see Chapter VII, p. 223. 
6 Uhlenbeck (AA *37) refuses to identify element B, or to call it specifically Caucasic. 
Nehring, however (Nehring, A., WBKL, vol. 4, 1936, pp. 7-229), feels certain that B 
is one of the group of which Caucasic may form a part. 


economy of the early Indo-European speakers was based were referred to 
in words derived mainly from element B. Copper and gold were known, 
and the words for these commodities come from Mesopotamia. 

Somewhere in the plains of southern Russia or central Asia, the blend- 
ing of languages took place which resulted in Indo-European speech. 
This product in turn spread and split, and was further differentiated by 
mixture with the languages of peoples upon whom it, in one form or other, 
was imposed. Some of the present Indo-European languages, in addition 
to these later accretions from non-Indo-European tongues, contain more 
of the A element than others, which contain more of the B. The unity of 
the original " I ndo- Europeans," could not have been of long duration, if 
it was ever complete. 

They split, perhaps very early, into two groups, designated by the treat- 
ment of the palatal explosives of the K group. Among one branch, the 
so-called Satem, this was changed to spirants (5); the other, called Cen- 
tum, preserved the original form of this sound, which also prevailed 
in the A or Finno-Ugric element. Centum speech became divided into a 
number of branches, of which surviving members are Keltic, Germanic, 
Italic, and Hellenic; Satem includes Slavic and Baltic, Armenian, Indie 
and Iranian, and probably Thracian, 6 in the sense of a contributing factor 
in modern Albanian. Others, such as Ligurian, Illyrian, 7 and Tokharian 
B (all Centum), have long been extinct. 

On the whole, the Indo-European languages have been spoken by 
people who combined agriculture with animal husbandry, who were 
organized into a patrilineal society with at least the germs of a differential 
class system, and who worshipped an Olympian pantheon of Gods. The 
initial formation of the Indo-European linguistic stock by blending does 
not antedate the age of metal; the common culture of the earliest Indo- 
European speakers, insofar as it existed as a unit, had much in common 
with those of both the peoples of the Aegean and Asia Minor on the one 
hand, and of central Asia on the other. The mythology of the Altaian 
Turks, for example, is so nearly identical with that of the early Scandina- 
vians that some close association in the not far distant past is necessary. 8 
Furthermore, the ritual of the horse sacrifice 9 is so integral a part of the 
religion of both Indo-European and Altaic-speaking peoples that recent 
diffusion alone cannot explain the identity. 

Indo-European languages as we know them must have come from east- 
ernmost Europe or western central Asia at no very remote time. Their 

6 Lowman, G. S., Language, vol. 8, 1932, p. 271. 

7 This may also be a factor in modern Albanian. 

s Chadwick, Nora K., JRAI, vol. 66, 1936, pp. 75-112. 

'Koppers, W., Anthropos, vol. 24, 1929, pp. 1073-1089; WBKL, vol. 4, 1936, 
pp. 279-411. 


spread over most of Europe, and subsequently over the western hemi- 
sphere, Australia, and large segments of Asia in which they were origi- 
nally not at home, is part of a general movement of expansion in which 
both race and culture have played their r61es. Yet we cannot with com- 
plete assurance associate any one culture earlier than the Iron Age with 
any specific form of Indo-European speech. Although Homer's heroes 
fought with bronze weapons, we are not sure exactly when and by what 
agency the pre-Dorian Greek dialects arrived in the racially and culturally 
composite Hellenic world; nor do we know exactly who brought Nasili 
speech to Asia Minor. 

One whole school of European archaeologists and linguists associates 
the Corded people with the diffusion of Indo-European speech. 10 Nehring, 
in a recent work of great detail and authority, would make the Danubians 
the original Indo- Europeans. 11 He would explain the Altaic cultural 
similarities by dividing the Indo-European culture and vocabulary into 
two elements: (1) an early horizon in which the ox was the most important 
domestic animal economically, and agriculture of primary importance; 
(2) a later horizon of indirect Altaic inspiration, in which the horse was 
supreme and agriculture secondary. 

At the moment the evidence is growing that certain forms of Indo- 
European speech were very ancient in more than one part of the Mediter- 
ranean basin. Whatmough has definitely identified Ligurian as Indo- 
European, 12 and Ligurian was very old in Italy and in the Rh6ne Valley. 
Sapir sees in Philistine a form of Indo-European; 13 and would make the 
ark of the covenant a spirit-placing on wheels like the portable wicker 
shrines of the later Mongols. But neither of these identifications need carry 
us back earlier in history than the time of the troubles in Mesopotamia at 
the end of the third millennium, when northerners caused restless nights to 
the Babylonian kings, and the Hyksos invaded Egypt. It was after these 
disturbances that the chariot first appeared in Libya; hence, the first 
southward burst of horse-nomads may have affected both shores of the 
Mediterranean, whatever languages they brought with them. 

The dates of the earliest certain appearances of Indo-European are 
about 1900 B.C., when the Nasili dialect which was incorporated into 
Hittite entered Asia Minor. The earliest Greek probably entered Hellas 
at the s^me time. About 1400 B.C., the ancestors of the Aryans of India 
were crossing the passes of Afghanistan into the Indus Valley, and some 
six hundred years later, their relatives the Iranian ancestors were founding 

10 That headed by Kossinna, who would likewise derive Indo-European speech from 
the Baltic. See Kossinna, G., Vr sprung und Verbreitung der Germanen. 
" Nehring, A., WBKL, vol. 4, 1936. 
12 Whatmough, J., The Foundations of Roman Italy. 
Sapir, E., JAOS, vol. 56, 1935, #2, pp. 272-281. 


the Persian empire. From roughly 1000-900 B.C. onward, as the earliest 
possible date, the bearers of the Hallstatt culture in central Europe were 
spreading the use of iron, and the Hallstatt people almost certainly spoke 
lllyrian. In Italy, the Villanova people were without reasonable doubt 
diffusing Italic speech in the peninsula, while some forms of lllyrian were 
introduced by a number of peoples, among whom were probably the 

All of these Indo-European speakers, from 900 B.C. onward, were asso- 
ciated in some way with the diffusion of iron metallurgy from a center 
which is still to be determined. The most commonly proposed location 
is northern Anatolia and the Caucasus; u whatever the history of the 
diffusion of Indo-European speech in the past, with the advent of iron, 
certain branches of it seem to have spread with great rapidity. The 
Hallstatt period in central Europe was followed by that of La Tene, the 
Late Iron Age, which lasted from 500 B.C. to the time of Christ; and this 
was the period of Keltic expansion and Keltic dominance, earlier than 
but parallel to the spread of Roman power and of Latin in the Mediter- 
ranean. After the phenomenal and immoderate scattering of the Kelts, 
who were destined to survive linguistically only on the western European 
fringe, far from their center of dispersion the Germanic peoples began, 
in the days of the Roman empire, their swelling and pushing, from Den- 
mark, southern Sweden, northern Germany, Holland, and the Norwegian 
coast. This reached every country in Europe and also North Africa. Un- 
like the spread of the Kelts, it was to achieve, in many quarters, linguistic 
and cultural permanence. 

The expansion of the Germans was followed by that of the Slavs, the 
youngest of the Indo-Europeans to effervesce in an orgy of numerical 
increase and of migration. This took place in full historic time, in the 
seventh and eighth centuries of our era, but, unfortunately, the light of 
history was dim in the part of Europe in which most of their expansion 

The foregoing digression into the field of comparative linguistics has a 
direct bearing upon the problem of the racial complexion of present day 
Europe. While it is not our primary purpose to discover the physical type 
or types of the undivided Indo-European ancestors, if they were ever 
actually undivided, it will be possible to find the common racial denom- 
inator, homogeneous or mixed, of the Iron Age spreaders of Indo- 
European speech and the accompanying cultures over Europe and parts of 
Asia. Once we have isolated the common factor, we may hope to locate 
its position in the roster of racial types previously known to us for it 
must have been some type or types with which we have already become 

" Wainwright, G., Antiquity, vol. 10, 1936, pp. 5-24. 


familiar in the earlier part of our study, and not a deus ex machina con- 
jured up by linguists and politicians. 


In beginning our survey of Iron Age Indo-European peoples, it may 
be well to choose the earliest instance in which we can definitely identify a 
language with a culture and a racial entity. This is true of the so-called 
Hallstatt culture associated with the Illyrian branch of Indo-European 
speech. Although usually classified with Centum, Illyrian, like Tokharian 
B, belonged to an ancient form of Indo-European which perhaps ante- 
dated the clear segregation into Centum and Satem. 15 

This culture arose in central Europe, with southern Germany and 
Austria as a focus, sometime shortly after the beginning of the first millen- 
nium B.C. It developed out of local Bronze Age origins carried over from 
the Urnfields, and in turn from Aunjetitz. Other Middle and Late Bronze 
Age influences reached it, particularly that of the tumulus culture of the 
south German highlands; likewise both cremation and the use of iron were 
introduced from outside. Still, whatever the complexity of archaeological 
detail, the Hallstatt civilization may be considered primarily the work of 
the indigenous central European population, with little if any accretions. 

The Hallstatt culture spread in many directions, including the south- 
east, where it penetrated Bosnia, and eventually Albania. It moved slowly 
northward, until it reached the Scandinavian and North German area, 
bringing iron to these regions relatively late; while to the southwest, it 
crossed France and penetrated Catalonia. To the immediate south, it 
likewise spread over the Alps into Italy, where the invading Illyrians split 
into a number of local tribal groups, including the Veneti. It would be 
foolish to claim that every site with Hallstatt cultural remains carries the 
bones or ashes of Illyrian speakers. This may only with certainty be as- 
serted for the central area, and for the regions immediately adjacent, while 
in the west it is fairly certain that some of the peoples in a Hallstatt level of 
culture were actually Kelts. 

The Hallstatt crania from Austria, including those from the type site 
itself, form a reasonably homogeneous, entirely long-headed group. 16 
(See Appendix I, col. 32.) This group is the legitimate, local successor to 
the Aunjetitz, and like the latter it resembles the Danubian Neolithic 

16 Whatnaough, J., The Foundations of Roman Italy > p. 177. 

16 Through combining several series, 24 adult male crania may be assembled. 

Hochstetter, F. von, MAGW, vol. 7, 1878, pp. 297-318. 

Rosensprung, L. M., MAGW, vol. 66, 1936, pp. 338-344. 

Schliz, A., AFA, vol. 37, 1910, pp. 201-251. 

Schurer von Waldheim, Hella, MAGW, vols. 48-49, 1919, pp. 247-263. 

Weisbach, A., MAGW, vol. 18, 1888, pp. 51-52. 

Zuckerkandl, E., MAGW, vol. 13, 1883, pp 89-118. 


series in many respects. In certain characters, however, it leans in a 
Corded direction, and these include a heightening of the orbits and a 
narrowing and lengthening of the nose. Certain of the individual crania 
are of definitely Corded type. Morphologically, as well as metrically, 
most of these skulls may without difficulty be designated as "Nordic 55 ; the 
browridges are moderate, the foreheads moderately sloping, the occiputs 
protruding, the parietals flattened, the malars compressed, the mandibles 
deep. The stature was apparently moderately tall. 17 

The Austrian Hallstatt series has close connections in two directions: 
first, with the local Bronze Age and Neolithic populations of central 
Europe which preceded it, and second, with the Germanic "Reihen- 
graber" people who followed it after a Keltic interruption. The similarity 
between Hallstatt and Germanic crania is a commonplace; and if the 
Reihengraber people were "Nordic," as is generally conceded, then so, in 
all likelihood, were the Hallstatt people. 

The significance of this double continuity is great. It traces the Nordic 
racial type, in its skeletal form, back to the Early Iron Age, and derives 
this with little alteration from the preceding Age of Bronze. The Bronze 
Age population which was thus the ancestral Nordic one was in turn de- 
rived from a mixture between the local Danubian Neolithic people, who 
came from the east, and the later Corded invaders. The complexity of 
the Middle and Late Bronze Age, therefore, and the disturbances caused 
by the introduction of cremation, during the latter part of the epoch, did 
not interrupt the racial continuity of central Europe, where racial move- 
ments, during the Late Bronze Age, seem to have been somewhat simpler 
than those of culture. 

Let us return to the specific problem of the Illyrian racial composition. 
So far, we have been dealing entirely with the Hallstatt remains from 
Lower Austria. The Hallstatt cemetery itself dates from the middle and 
later thirds of the period; but the neighboring Early Hallstatt site of Stat- 
zendorf, from which a series of five crania have been taken, contains noth- 
ing but long-headed examples, and these are the same as those from the 
type site itself. So the Hallstatt site is racially typical of the entire period. 

When we move to southern Germany, however, which was equally 
involved in the development of this culture, we find no such racial uni- 
formity. Crania from Wurttemburg, Bavaria, and the Bavarian Palatinate 
include, with the usual Austrian Hallstatt type, a large minority of brachy- 
cephals which may be considered as survivals from the Bronze Age. 18 

J7 Matiegka, H. (MAGW, vol 41, 1911, pp. 348-387), fails to segregate Hallstatt from 
Aunjetitz long bones, implies that both are the same, with a mean stature of 168 cm* 
18 Schliz, A., AFA, vol. 37, 1910, pp. 202-251. 
Schultz, B. K., VGPA, vol. 3, 1929, pp. 5-12. 


These include both planoccipital crania of the original Bell Beaker type, 
and a curvoccipital brachycephalic type which shows a Borreby relation- 
ship. It would appear, then, that in southwestern Germany, Hallstatt Nor- 
dics had invaded the region and had mixed with the Bell Beaker Binaries 
and the old Borreby sub-stratum. 

A large series from the Spreewald, situated to the north of this area 
and on flat land, consists entirely of purely dolichocephalic crania of the 
regular Austrian Hallstatt type, 19 which was apparently at home in the 
lowlands of central Europe, but not in the highlands, which had already 
given shelter to a tenacious brachycephalic population. In Bohemia and 
Silesia, as one would expect, Schliz finds typical Hallstatt dolichocephalic 
forms in small collections from each of these regions. One out of five Bo- 
hemian crania was brachycephalic, and none in a series of four from 

The generalization announced in the preceding paragraph applies like- 
wise to Switzerland, where the Hallstatt culture, like that of the Bronze 
Age, penetrated slowly, while the older economy and technique which 
had survived in part from the Neolithic persisted in large measure. Both 
long-headed skulls and those of brachycephals are found, as is to be ex- 
pected. In the available Swiss Hallstatt material, the majority of crania 
are brachycephalic. 20 

Let us turn southeastward and follow the Dinaric Alpine chain in the 
direction of the Balkans. In the mountainous section of southern Austria, 
the Hallstatt Nordic type is in the minority. Out of six skulls from Carniola, 
three are round headed and one is mesocephalic. The brachycephalic 
types seem without question to be predominantly Dinaric. In Croatia, 
however, seven adult skulls are all long headed, of the usual Hallstatt type, 
while two infantile skulls show brachycephaly. 

In Bosnia, we come to the famous site of Glasinac, 21 where a compar- 
atively large series of relatively late Illyrian remains contains again a 
mixture of types. The majority of the skulls are long headed and these 
show the same mixture of Danubian and Corded elements which we have 
already seen at Hallstatt itself. A few of the individual crania are very 
large, and reproduce the Corded prototype quite accurately. The brachy- 
cephalic skulls, although in the minority, are numerous enough to permit 
one to determine their racial affiliation with some accuracy. Almost all 

19 G6tze, A., PZ, vol. 4, 1912, pp. 264-350. This cemetery, unfortunately, was used 
at two periods; from 1000 to 500 B.C. when it was a Halls^att graveyard, and from 
500 A.D. on, when it was occupied by Slavic Wends. It is impossible to state how many 
of the crania belong to the Hallstatt people, and how many, if any, to the Slavs, but in 
either case the series represents one unified physical type of Hallstatt affinity. 

20 Schlaginhaufen, O., VNGZ, vol. 79, 1934, pp. 220-270. 
Weisbach, A., WMBH, vol. 5, 1897, pp. 562-576. 


belong to what might be called a modern Dinaric racial type. The skulls 
are moderately large with flattened occiputs, straight side walls, rather 
broad foreheads, and a very prominent nose, in the one instance in which 
the nasal bones were preserved. 22 The jaws are very broad with an ex- 
cessive bigonial diameter, but not noted for their depth. 

Metrically, these brachycephalic crania resemble the Bronze Age series 
from Cyprus, but are, on the whole, a little larger. They fall, as a matter 
of fact, into an intermediate position between the Cyprus series and the 
Bell Beaker group from the upper Rhineland, but in morphology are iden- 
tical with both. There is no doubt that we are dealing in this instance 
with a form of Dinaric which anticipates the modern population of Bosnia. 

This is the first occurrence of crania of this type in the Dinaric Alpine 
region in any considerable numbers. We have already seen, however, 
that this same type had entered these mountains by the beginning of the 
Bronze Age, in connection with the eastward movement of the Bell Beaker 
peoples. The round-heads at Glasinac and in Carniola may have been the 
descendants of these Bell Beaker refugees. It is also possible that this 
racial type may have been reenforced by migrations from the southeast, 
but there is no archaeological evidence to favor such a theory. 

As the Illyrians spread southwestward along the Dinaric Alps into 
Montenegro and Albania, they apparently blended with an indigenous 
brachycephalic mountain population which may have been more numer- 
ous than the invaders; for, with some additions and modifications, it per- 
sists as a predominant element today. In a small series of early Christian 
crania from a site near Split on the Dalmatian coast, 23 both Dinaric 
brachycephals and a few long-headed crania are represented. In Albania, 
a country which is almost completely unknown archaeologically, a single 
skull which belonged to a Romanized Illyrian group has been found in 
an Iron Age site in the tribe of Puka. 24 This skull is mesocephalic, and 
seems, insofar as we may judge, intermediate between the Illyrians of the 
old type and Dinarics. 

The significance of our study of the Illyrian peoples is as follows: on the 
plains of south central Germany and Lower Austria, where the Hallstatt 
culture arose, the racial type involved was skeletally a Nordic one. By 
this term we must understand that the Illyrian central type was similar in 
crapial dimensions, proportions, and general form to that of the Germans 
of the Volkerwanderung period. Historical evidence as to the pigmen- 
tation of the Illyrians is conflicting, 25 and insufficient to warrant the 

22 In all of the Glasinac crania the facial bones are missing 
2 Horvath, A., MAGW, vol. 36, 1906, pp. 239-248. 
* Lebzelter, V., AFA, vol. 45, 1919, pp. 143-146. 
" Lebzelter, V,, MAGW, 1929, vol. 59, pp. 61-126, 


formation of an opinion on this matter. This "Nordic" type is no special 
or separate race, but merely a variant of the larger Mediterranean family, 
of an intermediate metrical position. 

It finds a ready prototype in the Bronze Age population which stretched 
from Austria to Siberia, and which was in turn the product of mixture 
between Danubian peasants and Corded invaders. It seems most likely 
that the Illyrians were largely the descendants, more specifically, of the 
Aunjetitz people, through an Urnfields medium, or of some similar 
physical blend composed of identical racial ingredients. 


One of the most controversial subjects in the whole of European history 
is the physical composition of the Keltic peoples. The name Keltic has 
been applied to many racial types, real and imagined, from short, brunet, 
round heads to blond brachycephals and Nordics. Many modern prehis* 
torians take the stand that the Kelts were everywhere a small minority of 
aristocrats and conquerors, and that no special racial type accompanied 
their expansion in Europe. This position, however, becomes invalid when 
we examine the actual skeletons of Keltic speakers. There was a Keltic 
physical type, which the Kelts carried to their primary areas of coloniza- 
tion, and which will be described shortly. 

Although earlier identifications, however likely, are still questionable, 
we may state that the Kelts as such first appeared in the European historical 
setting about the year 500 B.C., with the beginning of the La Tne civiliza- 
tion. The home of the Kelts, or at least the country in which they devel- 
oped this brilliant Iron Age culture, lies without reasonable doubt in south- 
western Germany, in the upper drainage of the Rhine, 26 a country which 
had formed the western section of the original Hallstatt area. The eastern- 
most outposts of the early Keltic domain were Bohemia and Galicia, 
while, on the west and south, it touched the territory of the Ligurians and 
of the Rhaetians. The Kelts, therefore, were situated northwest and west 
of the Illyrians proper, and south of the Germans, who at the time were 
confined to Scandinavia and northwesternmost Germany. 

The Keltic languages are very closely related to the Italic group, of 
which Latin was a derivative. The period in which the Keltic languages 
became differentiated from other forms of Indo-European speech must, 
therefore, be as old as the departure of the ancestors of the Italici for Italy, 
and therefore must lead back to the Bronze Age. 27 Keltic, like Italic, is 
divided into two branches P-Keltic and Q-Keltic. |t is considered likely 

26 Hubert, H., The Rise of the Celts, p. 147. 

27 Although one school of Italic scholars derives the P-Italici from north of the Alps 
.n Iron Age times, all admit the Bronze Age dating of the Q-Italic arrival. For the de- 
tails of this controversy, see Whatmough, J., The Foundations of Roman Italy. 


that the phonetic separation which split both of the linguistic groups took 
place independently in each, and that the tendency for such a division was 
inherent in both Keltic and Italic at the time of their separation from one 
another. We do not know at what time the Goidelic or Q-Keltic dialect 
split off from the Brythonic or P dialect, but this cleavage again must have 
occurred at a reasonably early period, since the division was complete at 
the time of our earliest knowledge of these languages. Q-Keltic has sur- 
vived only in Ireland, Scotland, and on the Isle of Man. All other known 
dialects, living and extinct, from Asia Minor to Wales, have been of the 
P variety. 

The Keltic expansion, which began about 500 B.C., was a rapid and 
extensive one. The Kelts were an extremely mobile people who conquered 
and wandered far, and at the time of their expansion were apparently 
numerous as well. Their well-known migrations carried them over the 
Alps into Italy, down into southeastern Europe where they invaded 
Greece, and even over into Asia Minor where they established the short 
lived Galatian colony. Their main expansion, however, lay to the west. 
Belgium and northern France became great Keltic centers, from which 
some of them migrated down into northern Spain. This westward move- 
ment carried them also into the British Isles, where the Q-Keltic people 
settled Ireland, and their P-Keltic brethren established themselves in 
England and Wales. Large sections of Scotland were to remain free for 
the most part from these Keltic invaders until after the time of Christ, 
when the Goidels crossed over from Ireland. 

The question as to the linguistic identity of the previous inhabitants, 
the Picts, is an open one. At present, the tendency is to consider them, and 
the pre-Goidelic Cruithni of Ireland, as speakers of some early form of 
Keltic. The further question as to whether or not the Goidels crossed Eng- 
land in their journey to Ireland is likewise open, but the prevailing tend- 
ency is to bring them over the old sea road from northern Spain, which 
they had previously entered by way of France, and to deny that they so- 
journed in England at all. 

In their period of development in southwestern Germany, the relation- 
ship between the Kelts and Illyrians must have been intimate, for the 
Kelts received iron from a Hallstatt source, and were actually, during the 
Early Iron Age, participants in a Hallstatt form of culture. The major 
factor which served to differentiate La T&ne from Hallstatt culture was 
the incorporation, by the former, of many elements derived from the clas- 
sical Mediterranean world. The Kelts were situated at a favorable spot 
for the reception of such influences; Greek influences moved up the Rh&ne 
and Sadne from Marseilles, while those from Rome crossed the Alpine 
passes into Bavaria and Switzerland and thence into the Keltic homeland, 


In addition to the Hallstatt Iron Age base and classical accretions, we 
must further acknowledge the influences of some eastern European grass- 
land culture, for the Kelts rode astride as well as in chariots, and the 
P-Kelts introduced trousers to western Europe. This garment was central 
Asiatic in origin, and was typical of the Scyths, whose period of cultural 
efflorescence in the east was contemporary with and parallel to that of the 
Kelts in the west. Philologically, there are a number of close linguistic 
connections between the Kelts and the Indo-Iranians, which may reflect 
this or an earlier cultural contact. It is most likely, however, that the 
principal contact between the Keltic-speaking peoples and the Iranian 
horsemen of the eastern European plain took place during the early years 
of the great Keltic expansion. 

Turning back from Keltic expansions to Keltic origins, we find no 
cultural disturbances in southwestern Germany which would permit the 
arrival of the Kelts from elsewhere between the Hallstatt epoch and the 
early La Tene. Before the Hallstatt, however, the spread of the Late Bronze 
Age Lausitz culture into this region from eastern Germany may conceiv- 
ably have brought a large number of people, impossible to identify because 
of their practice of cremation. These people may well have been the 
bearers of Keltic speech. Since the related Italici were themselves Urn- 
fields cremators before they succumbed to indigenous burial rites in Italy, 
this identification is rendered more than likely. Hubert has, indeed, postu- 
lated an earlier Ligurian-speaking population in the Keltic cradle-area. 28 
The derivation of the Kelts from a Hallstatt cultural horizon, in part of 
the earliest region of Hallstatt development, while the main current of 
Hallstatt cultural expansion was borne by Illyrian speakers, seems incon- 
gruous. One must remember, however, that the Nordic skeletal type with 
which the Illyrians were identified in Lower Austria was confined, in its 
purely dolichocephalic form, to the lowland country north of the Bavarian 
foothills, while the Keltic area of development was, in its strictest limits, 
within the highland zone. Here the Kelts developed their own culture 
independently of the Illyrians and retained their own language. 

Keltic cranial material from the southwest German center of Keltic 
development is surprisingly scarce. Schliz has described six skulls, and 
notices of three others have appeared in more recent publications. 29 Of 
these nine, one is dolichocephalic, four are mesocephalic, and four are 
brachycephalic. Although this small group is far from sufficient to dis- 
close the racial type of the Kelts in their homeland, it is enough to show 
us that a round-headed element played a considerable part in the develop- 

28 Hubert, H., The Rise of the Kelts, p. 159. 

29 Jacob, G., AFA, vol. 20, 1891-92, p. 181. 
Ortmann, R., JVST, vol. 15, 1927, pp. 56-59. 
Schliz, A., AFA, vol. 37, 1910, pp. 246-251. 


ment of this ethnic group. The brachycephals involved are large headed 
and powerfully built, with long faces, and rather high orbits; the foreheads 
are sloping and only slightly bowed at the junction of the facial and cranial 
planes. The inference is that these brachycephals were derived from the 
older combination of Bell Beaker and Borreby types which was formed in 
the upper Rhine country at the beginning of the age of metal, and which 
persisted into the Hallstatt period. These seem to have mixed with the 
expected intrusive Nordics. We must really wait until we examine larger 
series of Keltic crania from elsewhere, however, before passing judgment 
on the final result of this blend. 

A better picture of the La Tene type may be obtained from the study of 
its early eastern extension. Hellich's series from Bohemia 30 (see Appendix 
I, col. 33) is the only single group of central European La Tene crania of 
any consequence. This includes 27 male crania, most of which are dolicho- 
cephalic, but which contain a significant minority of brachycephals. In 
general, the La Tene skulls are not in any important metrical way distin- 
guishable from those of the preceding periods of which we have clear 
knowledge that is, Aunjetitz and Hallstatt. They represent merely a sub- 
variety of the same general combination of types, with a brachy cephalic 
accretion which makes the total series rnesocephalic. 31 But there are other 
features, however, which render them as a group slightly different; the 
vault has a tendency to be low in proportion to its breadth, and the upper 
face is long in proportion to the total face, for the Keltic jaw, although 
broad at the gonial angles, is not as deep as that of other Iron Age Nordics. 
A composite series of eleven male crania from the type site of La Tene on 
Lake Neufchatel in Switzerland, and nearby burial places, 32 is almost, 
exactly the same as the Bohemian series; the vaults of the Swiss La Tene 
people, who may in part be identified with the Helvetii, are even lower 
than those of the Bohemians. As one might expect, the Swiss series con- 
tains a number of high brachycephals, with cranial indices as high as 
90; 33 but on the whole, most of the few Kelts whose remains have been 
studied in Switzerland were no different from those in Bohemia. 

Less than a dozen skulls serve to identify the Keltic racial elements in 
Austria and in the Dinaric Alpine mountain zone. 34 On the whole, this 

80 Hellich, B., Praehistoricke lebky v &chach & Sbirky Musea Krdlovstvi tfesktho. 
31 Schliz's series of 14 crania from Bohemia, 3 from Moravia, and 2 from Silesia do 
not differ from those measured by Hellich. Schliz, A., AFA, vol. 37, 1901, pp. 246-251. 

82 Virchow, R., ZFE, vol. 16, 1884, pp. 168-181; ibid., vol. 18, 1886, pp. 561-566. 
Lagotala, H., BMSA, ser 7, vol. 3, 1923, pp. 4-9. 

83 Schlaginhaufen, O., AFSA, N. F. Bd. 38, 1936, pp. 226-236. 

84 Poch, H., MAGW, vol. 56, 1926, pp. 255-270. 
Lebzelter, V., WPZ, vol. 22, 1935, pp. 104-105. 
Luschan, F. von, MAGW, vol. 8, 1879, pp. 85-89. 
Schliz, A., loc. at. 


evidence is not satisfactory, but it serves to indicate that the regular 
mesocephalic type and one or more types of brachycephals were present. 
The most southeasterly Keltic skull known is one from Kupinovo, near 
Belgrade in Serbia, which belonged to a Dinaric brachycephal similar to 
those found at Glasinac, and this again witnesses the persistence of this 
Dinaric element during the Iron Age in or near the modern Dinaric 

Before turning to the abundant remains of the Kelts in France and the 
British Isles, it may be well to review what evidence we have for their 
racial type in central Europe. Here the Kelts seem to have been a com- 
posite people, a blend of the different brachycephalic elements left over 
from the Bronze Age in the mountainous zone of southern Germany, and 
invaders of Nordic type from the plains to the north and east. One sup- 
poses that the Keltic linguistic element came with the later group. 

Sculpture from Greece and Rome gives us a picture of the living Kelts who 
reached the lands of classical civilization by eastward and southward move- 
ments. The well-known Dying Gaul and similar statues show a strongly 
muscled type with mesocephalic or brachycephalic head form, a rather 
short face with a square jaw, a straight and rather prominent mesorrhine 
nose, with horizontal or elevated tip and full nostrils, heavy browridges, 
a broad forehead, and stiff, bristly hair. This type, while familiar enough 
in western Europe, is not one which accords with the majority of the Keltic 
skeletons. The typical Keltic face was long in the upper portion, shallow 
in the mandible, long and narrow of nose, often with a convex profile, and 
the forehead was extremely sloping and the vault low. This has its most 
frequent counterpart today in the British Isles. While the type selected by 
the classical sculptors to represent the Kelts must have had its living 
models, these may have been drawn from the brachycephalic minority. 

Most of the La T&ne material from France comes from the north, from 
the Marne region, where the Keltic settlement seems to have been par- 
ticularly strong. Fortunately, large and competent series of the Gauls of 
this district, before and after the Roman conquest, furnish adequate in- 
formation. 36 (See Appendix I, col. 34.) Both groups are alike, showing 
that submission to Roman rule did nothing to change the physical type 
of this particular people. 

The Gauls as so represented were mesocephalic, mesoprosopic, and on 
the upper borders of leptorrhiny. The vault, as with all characteristic 

La Tne Keltic groups, is not distinguished for its height, and in the large 


** Raymond, P., RP, vol. 2, 1907, pp. 10-22, includes 20 males. 

Wallis, Mrs. Ruth Sawtell, unpublished measurements in Musee Broca, cole 
d'Anthropologie, and Mus6e d'Histoire Naturelle. Includes 28 pre-Romans and 83 
Gallo-Romans, all males. 


and more reliable post-Roman series, it is definitely low. Like their rela- 
tives in central Europe, these Gauls were not noted for tall stature; a 
mean of 166 cm. is only moderate. 

In other parts of France, the Keltic racial continuity was of variable 
intensity; in Lorraine and Beaune, 36 the usual type was found; but in 
Haute Savoie and Vend6e the earlier brachycephalic population is strongly 
represented in Keltic tombs, 37 while out on the tip of Brittany, Neolithic 
survivors of Mediterranean type, with perhaps some Gaulish admixture, 
persisted until the period of Roman conquest. 88 Only in the north, there- 
fore, did the Kelts make a firm imprint in the early population of what 
was to become the French nation. 

The Kelts in the British Isles are known to us by a large series of Bry- 
thonic crania from England and southern Scotland, assembled by Morant 39 
(see Appendix I, col. 35) ; these are three millimeters longer headed than 
the Bohemian and Swiss series, but nearly identical in vault dimensions 
with the French; facially they are the same as all of the others. Smaller 
collections of Goidelic crania from Ireland show the skulls from this coun- 
try to be exactly the same as those from Great Britain. 40 Several morpho- 
logical features distinguish these skulls, of the typical, or mesocephalic, 
group which in the British Isles seems largely to lack the brachycephalic 
minority which accompanies the main type in central and eastern Europe. 
The forehead is quite sloping; the vault, when seen from behind, gives a 
cylindrical impression, rather than that of a rhomboid or rectangle, as 
with other Nordic crania. The upper face is quite long, the mandible 
wide at the back, and relatively shallow. The nose is often very prominent. 

The skeletal material from Ireland (see Appendix I, col. 36) is not 
numerous enough to permit regional studies, or other statistical niceties; 
but in Great Britain there are, on the contrary, a number of local series 
sufficient to show that the racial complexion of that island was not, during 
the Iron Age, completely uniform. One of these, that of the erroneously 
named "Danes' Graves" at Driffield, Ygrkshire, 41 containing 29 male 
crania, is identical in every known respect with the Aunjetitz skulls from 
central Europe a pure (if the adjective pure may be used of a composite 
type) Hallstatt or Nordic local population; purely dolichocephalic, in 
contrast to the usual Keltic mesocephaly; and relatively high- vaulted, 

86 fiamy, E. T., Anth, vol. 17, 1906, pp. 1-25; vol. 18, 1907, pp. 127-139. 
^Baudoin, Marcel, BSAP, vol. 6, 1912, pp. 321-346. 

38 Vallois, H. V., Les Ossements Bretons de Kerne, Toul-Bras, et Port-Bara. 

39 Morant, G. M., Biometrika, vol. 18, 1926, pp. 56-88. Also Hooke, Beatrix, and 
Morant, G. M., Biometrika, vol. 18, 1926, pp. 99-104. 

40 Martin, G. P., Prehistoric Man in Ireland, Twelve Iron Age skulls are listed. 

41 Wright, W., JRAI, vol. 33, 1903, pp. 66-73; Archaeologia, vol. 60, 1906, pt I, 
pp. 313-324. 

Mortimer, J. R., Man, vol. 9, 1909, pp. 35-36. 


again non-Keltic, although the stature, 167 cm., is presumably no different 
from that of the Kelts, 42 

It is impossible to derive this group from the local Neolithic, which was 
noted for its extreme absolute cranial length; or from the dolichocephalic 
element of the Bronze Age, which was again larger, longer, and higher 
skulled; it resembles not only the earlier Aunjetitz and Hallstatt, but also, 
although to a lesser degree, the contemporary Scandinavian Iron Age 
people in the period immediately before the Germanic Volkerwanderung. 
All of the archaeological material found in the Danes' Graves has never 
been satisfactorily identified. 48 Although the dominant Keltic tribe of 
that neighborhood, the Parisii, seems culturally represented, it is unlikely 
on archaeological as well as on racial grounds that the majority of the 
men buried in these graves came from the Marne, whence the usual Bry- 
thonic tribes migrated to England. Two of the fibulae found in the scanty 
remains have Scandinavian affinities; despite this clue, however, we must 
leave open the question of the immediate origin of the Danes' Graves 
people, and render the verdict: "Central European Nordics found in 
Yorkshire during the late Iron Age, provenience unknown." 

Another local group which shows aberrant tendencies is that of eleven 
male crania from Berkshire, of which the length, breadth, and circum- 
ference alone are available; 44 the figures are 193.3 mm., 149.6 mm., and 
552.2 mm. The cranial index is 77. These mesocephalic crania are so 
much larger than those of the total Iron Age population that some other 
origin must be postulated. One recalls the extravagant dimensions of 
both Neolithic and Bronze Age crania in England, and may only suppose 
that this local group represents a relatively unaffected survival. Since both 
Bronze Age and Neolithic racial types may be picked out of any moderate- 
sized gathering of living Englishmen, or of their transatlantic relatives, it is 
not surprising to find a few in Berkshire during the Iron Age. 

The descriptions of the Kelts, in Britain, in France, and in other parts of 
Europe, at the hands of classical authors, give us a definite picture of their 
pigmentation. Blondism was by no means characteristic of the Kelts as a 
whole. Rufosity was common, and the hair color was essentially mixed. 
Caesar himself noted the contrast between the ordinary Gauls and the 
partly Germanic Belgae, to whom he had to turn to find real blonds for 
his triumph. Furthermore, the Romans noted the Keltic practice of 
bleaching the hair to simulate a blond ideal, as in Greece. 

42 We know the stature of Kelts in the British Isles only from^a small Irish group, and 
by inference from comparison with mediaeval English counterparts of Iron Age skele- 

48 Greenwell, W., Archaeologia, vol. 60, part 1, pp. 251-312. 

Bremer, W. Real, vol. 1, pp. 229-230, article " Arras.'* 

44 Morant, G. M., Biomctrika, 1926, vol. 18, pp. 56-98. 


On the whole, the Kelts were a mixed group in race as in culture; their 
ancestry includes both long heads of some central European Nordic type, 
which was in turn a combination of several Mediterranean sub-types, and 
brachycephals from the region in southwestern Germany in which the 
Binaries of Early Bronze Age introduction had blended with earlier round 
heads of Mesolithic origin. Out of this combination, the Kelts developed 
an easily identified national type, of considerable constancy, which was 
to be of some importance in the world, especially in Britain and the na- 
tions derived from her* 


Before proceeding to study the rest of the Iron Age Indo-European 
speakers in their homes north of the Alps, let us examine the racial posi- 
tion of those near linguistic relatives of the Kelts, the Italici, who lived 
south of that barrier, and who played a r61e of the utmost importance in 
the history of Indo-European speech. The racial problem in Italy is 
nearly as complicated as in Greece, but the recent work of Whatmough, 
paralleling that of Myres, makes its solution equally possible. 45 

We have already witnessed the accretion of various racial elements in 
Italy up to and through the Bronze Age. To a Neolithic Mediterranean 
sub-stratum were added tall, long-headed Megalithic invaders who came 
by sea, and Dinaric brachycephals from the eastern end of the Mediter- 
ranean. In the Late Bronze Age, Urnfields people crossed the Alps from the 
north, and settled in northern Italy. Some of them built the terremare 
settlements in the Po Valley, while their descendants or others like them 
were responsible for the Villanova settlements in the Bologna region, and 
similar sites as far south as Latium. These collective Urnfields peoples 
came from central Europe, rather than from the nearer Swiss center. The 
Italic languages, like Keltic, were without reasonable doubt introduced 
by the Urnfields people. Like Keltic, they split into P and Q forms, with 
Oscan and Umbrian as P, and Latin and Ealiscan as Q. Latin itself, in its 
historic form, was a mixture of Villanovan Italic plus Etruscan plus some 
altered Greek, plus early Mediterranean words, including plant names. 46 
The non-Italic accretions bear witness to the influences which met the 
early Romans, while its major Italic character throughout attests the per- 
sistence of the Romans in retaining the nucleus of their own speech through 
centuries of Etruscan overlordship. 

We know comparatively little about the racial composition of the early 
Italic people in pre-Roman times. Two crania from Remedello 47 are 

46 Whatmough, J., The Foundations of Roman Italy. 

Ibid., pp. 276-277. 

Zampa, R., APA, vol. 20, 1890, pp. 345-365. 


both those of dolichocephals of moderate size; one of them, which is cer- 
tainly a male, has a stature of 168 cm. Two early Romans ** were like- 
wise dolichocephals of the same size and proportions as many of the Nordic 
groups north of the Alps; while a third, from the pre-Republican cemetery 
of Corneto Tarquinia, which can be more accurately defined, resembles 
a small male series of eight Christian Roman skulls, dating from the first 
to fourth centuries A.D. 49 These nine male crania are identical metrically 
with the means for the La Tene Kelts in Bohemia, and the Gauls and 
Gallo-Romans of the Marne. The same mesocephalic, leptorrhine form 
is found in each case. 

Historically, the Romans should have been a mixture of Villanovan 
Italic northerners with Etruscans and Neolithic and Bronze Age predeces- 
sors. 60 The little crania material at hand points entirely in the northern 
direction, and confirms the relationship between Kelts and Italici, insofar 
as it may be used. On the other hand, the addition of Etruscan meso- 
cephals with Dinaric and Mediterranean elements would not greatly alter 
the early Kelt-like Italic metrical form. 

The early Romans, judging from the busts of their descendants in the 
days of Augustus, and of descriptions, were not very tall, as a rule, but were 
often of heavy bodily build. Their skulls were flattish on top, and rounded 
on the sides, like those of Kelts. The facial features included the well- 
known "Roman" nose, which may have been partly derived from an 
Etruscan source. On the whole, the well-known sculptures of Caesar, 
Augustus, and others, although not reliable from the standpoint of ac- 
curate measurement, indicate that a mesocephalic to brachycephalic 
head form was admired. Their facial type is not native to the Mediter- 
ranean basin, but is more at home in the north. Nevertheless, the 
Romans considered the Kelts who invaded Italy tall and blond; hence 
the blondism of the Romans, including rufosity, must have been in the 
minority. 61 

More detailed information may be obtained by studying the remains of 
Romans who died away from home in the colonial service of the empire. 
For example, an officer of the sixth legion, named Theodorianus, stationed 
at York, came from the small city of Nomentum, in Latium. Three others, 
also buried at York, were also native Romans. 62 These four were all of one 
type, and' very much alike: dolicho- to mesocephalic, with low vaults, low, 
broad foreheads, very aquiline noses, and short, broad, square faces. The 

Sergi, G., ARAL, Anno 280, 1883, 10 pp. > 

49 Moschen, L., Cram Romani delta Primer a Epoca Cristiana, 1894. 

Prdbstl, L. AFA, vol. 45, 1919, pp. 80-81. 

60 Whatmough, op. cit., p. 267. 

"Rochet, C., MS4P, vol. 3, 1868, pp. 127-145. 

62 Davis, J. B., and Thurman, J., Crania Britannica, 1865, Part II. 


skulls of two other pure Roman officers from Bath and Gloucester are the 
same, as is one from Lincoln. 68 

A group of eight male Roman crania from Rheinzajpern on the Rhine, 64 
belonging to real Romans from Italy, are the same as the individuals from 
Britain, and almost identical with the eight males from Rome itself of the 
Christian period, and the early Roman from Corneto Tarquinia. These 
scattered references from various quarters, although few, are so alike that 
we must conclude that the Romans, however mixed, had formed a charac- 
teristic local or national physical type, which was mainly of Italic origin, 
and closely related originally to the Keltic. 

The Italici, however, were not the only Indo-European speakers to in- 
vade Italy from the north. The Ligurians, of whom we have no certain 
skeletal remains, probably entered from Gaul, and may have been earlier 
than the Italici. On the eastern watershed of the Italian peninsula and in 
the Po Valley lived, in early protohistoric and historic times, various tribes 
of Illyrian speakers, notably the Veneti. To the Illyrian group may have 
belonged the people who buried in the cemetery of Novilara, on the cen- 
tral Adriatic coast, 65 about the eighth century B.C., contemporaneously 
with the Villanova people. The site belonged to a tribe called the Piceni, 
who in the seventh and sixth centuries developed a high culture and later 
declined, becoming subjects of Rome. 

The doubt as to their ethnic origin may be partly dispelled by a knowl- 
edge of their physical remains, A series of eighteen male and thirteen 
female skulls is homogeneously dolichocephalic, with the low mean male 
cranial index of 71.2; the skulls are high- vaulted, narrow- faced, and leptor- 
rhine. The series is very similar to those of Hallstatt Illyrians farther north, 
and the stature, 165.5 cm. for males, is tall enough to support this. 
Whether or not they spoke Illyrian, they were of Illyrian racial type, and 
the Illyrian invasion of northeastern Italy was undoubtedly a real one in 
the racial sense. 


What the Kelts were to western Europe, the Scythians and their rela- 
tives became, at about the same time, to the treeless plains to the east. 
Riding astride, wearing trousers, and sleeping in covered wagons, they 
spread rapidly over the grasslands of eastern Europe and western central 
Asia, shifting so adroitly that Darius with his army could not catch them, 
and disappearing almost as rapidly from the face of eastern Europe as they 
had appeared. Like the Kelts, they were both dazzling and ephemeral. 

M Browne, C. R., PRIA, vol. 2, ser. 3, 1899, pp. 649-654. 
M Probstl, L., AFA, vol. 45, 1919, pp. 80-81. 

88 Whatmough is in doubt as to their linguistic affiliation. Whatmough, J., op. a'/., 
pp. 202-205. 


But unlike the Kelts, their way of living, perfectly adapted to the grass- 
lands on which they roamed, was destined long to survive their identity 
as a people. 

About 700 B.C. the Scyths were first noticed in the lands to the north of 
the Black Sea. 66 Their domain reached from north of the Danube and 
east of the Carpathians across the fertile plains of eastern central Europe 
and southern Russia to the River Don. From this country they were sup- 
posed to have ousted the somewhat mysterious Cimmerians, Although 
the Don formed their eastern boundary, beyond it lived other groups of 
nomadic peoples culturally similar to the Scythians. These included the 
Sarmatians, their immediate neighbors to the east, who were, according 
to Herodotus, the result of a mass marriage of Scythian youths and Ama- 
zon maidens. The speech of the Sarmatians was said to be somewhat dif- 
ferent from that of the Scythians, owing to the inclusion of Amazon words 
and an Amazonian manner of pronunciation. Beyond the Sarmatians 
lived the Massagetae, and beyond them the Saka. The word Saka, how- 
ever, was used by the Persians as a general term, to include all of the no- 
madic peoples to the north of the Iranian plateau, in the two Turkestans. 

In costume, in weapons, in methods of transportation, in living quarters, 
and in the totality of material culture, these people formed a continuous 
cultural zone from the Carpathians to China. It has been the custom to 
consider the Scythians a people of Asiatic origin who developed this high 
and specialized form of pastoral nomadism in central Asia and brought 
it with them to eastern Europe. Proponents of this school have suggested 
that the Scythians were a mongoloid people, and that they employed 
some Altaic form of speech. Another school holds that they were Euro- 
pean in physical type, and spoke Iranian, while their cultural breed- 
ing ground lay somewhere to the east of the Caspian. 

We do not know what language the Scythians spoke, nor is it likely that 
its exact affiliation will ever be definitely established. Their geographical 
position, however, and their association with the ancient Persians, makes 
the Iranian hypothesis very likely. This theory is further strengthened by 
the study of the language of the Ossetes, a living people of the Caucasus, 
who are supposed, on historical grounds, to be descendants of the Alans, a 
branch of the Sarmatians. Their language is definitely Iranian. 

Although the general manner of living enjoyed by the Scythians does 
resemble in a remarkable degree that of the later Huns, Turks, and Mon- 
gols, one looks in vain for some of the cultural traits of these later Altaic 


M The sources for the historical and cultural portions of this section include Herodo- 
tus, book iv, ch, 59-75; Hippocrates, de Aere; Minns, E. H., Scythians and Greeks; Junge, 
J, ZFRK, vol. 3, 1936, pp. 68-77; and Wm. M. McGovern's work, The Early Empires 
of Central Asia, which was consulted in advance of publication. 


speakers which may be ascribed to a relatively recent Siberian origin. 
These include the yurt or collapsible felt-domed house, and the Turko- 
Mongol type of shamanism. The Turks and the Mongols, without ques- 
tion, took over almost completely the whole Scythian style of culture, 
but they added to it elements of their own which reflected their former 
habitat and manner of life. A few traits connect the Scythians with their 
neighbors to the north, the Finns; among these might be cited the sweat 

The Scythians proper possessed a type of feudal organization headed by 
a king, who ruled over four provinces each of which had local governors. 
These Scythian kings were all buried in a royal burial ground in the 
region called by the Greeks the Land of the Gerrhi, which was situated in 
the bend of the Dnieper River near Nicopol. No matter where the Scyth- 
ian monarch died, his remains would be deposited, in a funeral chamber, 
with great ceremony and with an extravagant quantity of human sacri- 
fice, underneath a huge mound erected for that purpose. The richness of 
the burials, and the wholesale suttee, are reminiscent of the ancient Sume- 
rians, and of the early Egyptians. The eventual Sumerian origin of this 
Scythian custom is not unlikely. 

This region of the Royal Scythian burying ground has been a source of 
great activity for both treasure hunters and archaeologists. The Scythians 
had a definite idea that this was the place in which their kings were natu- 
rally at home, and while it may not be wise to stress this point too much, 
it would seem that this location may have reflected their notions as to 
their original dwelling place, or at least that of their royal clan. Similarly, 
the Mongols in later times buried their dead in a restricted area in the 
Altai Mountains, which they considered holy ground. 

During the first century B.C., the Sarmatians penetrated westward, 
crossing the Don, and driving the Scythians from their former homes. 
About 200 A.D., the Goths took the Scythian country from the Sarmatians, 
and in turn adopted much of the Scythp-Sarmatian culture, becoming 
great horsemen and learning to live in wagons. The Alans were the only 
branch of the Sarmatians to retain their integrity in face of this Germanic 
onslaught. They built up a great kingdom between the Don and the 
Volga, reaching as far as the Caucasus, including in it most of northwestern 
Turkestan. Between 350 and 374 A.D., the Huns destroyed the Alan king- 
dom. Some of the Alans went westward with the Huns, others accom- 
panied the Vandals to North Africa, and a few, as previously mentioned, 
survive in the Caucasus as Ossetes. 

Although these Iranians (if the Scythians and Sarmatians really were 
Iranians) were replaced by Altaic speakers in southern Russia, and 
throughout the breadth of their Asiatic domain, this process took some 


time, and Iranian languages clung on for a long while in Kashgaria and 
in the oases of Russian Turkestan. Undoubtedly, the Scythians and their 
relatives were not destroyed, but were absorbed and reincorporated. 

In studying the racial type of the Scythians, one must remember that 
they were not considered a homogeneous group by Herodotus, who is our 
chief historical source. They consisted of an inner clan called the Royal 
Scyths or True Scyths, who were the nobles and leaders, and, as a second 
element, the whole group of nomadic tribes of which the Royal Scyths 
were the integrating force. Herodotus also makes it clear that the Scythians 
kept many slaves. Only the Royal Scyths refused to own slaves, but em- 
ployed youths of pure Scythian blood as bodyguards, and sacrificed these 

Redrawn from Minns, E. H., Scythians and Greeks, p. 201, Fig. 94. 

in their tombs. Thus, the Royal Scythian burial mounds must contain a 
relatively pure Scythian group. 

One must not imagine that the Scyths and their slaves were the only 
inhabitants of southeastern Europe during the last seven centuries before 
Christ and the first two of our era. Herodotus mentions the agricultural 
Scythians, who were probably some earlier sedentary people or peoples 
who remained as underlings of the Scythians and their providers of cereal 
food. We must remember that much of the Scythian territory had been 
farmed as early as Neolithic times. 

There can be little doubt, even before examining the skeletal evidence, 
that the Scythians and Sarmatians were basically if not entirely white men 
and in no sense mongoloid. The only definite description of them which 
we have from classical literature is that of Hippocrates, who called them 
white-skinned and obese, but this designation was employed by the father 
of medicine to prove one of his environmental theories. In later times, 
the Alans are described as having golden hair. 

Fortunately, we are not limited to literary references. The Scythians 
themselves, under the influence of powerful Greek colonies on the north 
shore of the Black Sea, and particularly in the Crimea, produced a dis- 


tinctive style of realistic art in gold repousse" e. These representations in- 
clude a number of portraits of Scythians in very realistic and life-like poses. 
They show a well-defined type of heavily bearded, long-haired men with 
prominent, often convex, noses. The browridges are moderately heavy, 
the eyes deep set. These faces are strikingly reminiscent of types common 
among northwest Europeans today, in strong contrast to those shown in 
the art of the Sumerians, Babylonians, and Hittites, which are definitely 
Near Eastern. The face, therefore, is definitely Nordic, while the body 
build looks often thick-set and very muscular, but this may be due to the 
clothing, which includes baggy trousers and jackets with full sleeves. The 
pointed caps which they wear and the long hair make it impossible to form 
a useful opinion of their head form, but this is unnecessary, since we may 
soon discover it from reference to the cranial material. Persian representa- 
tions of Saka show exactly the same type, depicted by the followers of an 
entirely different school of art, and hence this type cannot have been an 
unfounded convention. 

There is, in the anthropometric literature, sufficient data to permit the 
reconstruction of the Scytho-Sarmatian cranial type or types. The most 
extensive group, and that which may be used as a basic series, is DonicTs 
collection of seventy-seven Scythian crania from kurgans of Bessarabia, 
which was one of the favored Scythian pasture lands during the height of 
their domination. 57 (See Appendix I, col. 37.) The fifty-seven male crania 
of this series are not homogeneous, but fall into two types, a long-headed 
and a round-headed, with the former greatly in the majority. 

The means of these Scythian skulls show them to be low mesocephals 
of moderate cranial dimensions, but with a low vault height. The cranial 
means are, in fact, almost identical with those of the Keltic series from 
France and the British Isles. They resemble the Aunjetitz and Hallstatt 
skulls only as much as the Keltic series mentioned resemble these latter. 
They are, furthermore, metrically identical with the previously studied 
skulls from the Minussinsk region of southern Siberia, which may have 
been contemporaneous with them. 

One of the peculiarities of the Scythian skulls is a low mesene upper 
facial index, lower than that of the Kelts or of the Minussinsk people. 
Doniti has shown, however, that this low upper facial index is mostly 
asspciated with the brachycephalic element in the group, and the same 
is true of many of the chamaeconch and mesorrhine skulls. When the 
brachycephalic element is eliminated, therefore, one finds these skulls 
to be narrower faced, and narrower nosed, and to fit more nearly into 
a central European Nordic category. Other series of Scythian crania 
from southern Russia and from the Caucasus show the same general 

67 Donifci, A., Crania Scythica, MSSR, ser. 3, Tomul X, Mem. 9, Bucharest, 1935, 


characteristics as that of Donici's type series, but are in most cases 
purely dolichocephalic, which leads one to suppose that the brachyce- 
phalic element in the Rumanian skulls may have been at least partly of 
local origin. 68 

Other collections of Scythian crania vary in their mean cranial indices 
from 72 to 77. Those from the Kiev government, a Scythian center, have 
a mean of 73. 59 A series of eighteen Sarmatian crania from the Volga, 
although otherwise the same as the others, has a cranial index of 80. 3. w 
However, one hesitates to consider this typical of the Sarmatians as a 
whole, since both the Alans 61 and the early Ossetes 62 were long headed. 
The former preserved the original Scythian Nordic type until the ninth 
century A.D. 

Of especial interest is a rich kurgan in the Royal Scythian burial dis- 
trict, 63 near Alexandropol; this was one of the most imposing kurgans of 
Russia, not only for its size but for the quantities of gold placed with the 
dead king, and of animals sacrificed for his convenience. The kurgan con- 
tained five skulls in the primary interment; one of these was a large male 
of Corded type. 64 Another is a brachycephal with a vault especially wide 
behind, with a broad face and a narrow nose, resembling a Turkish or 
perhaps a Bell Beaker type; two are narrow skulls of the normal Scythian 
Nordic variety, while the fifth, that which occupied the king's chamber, 
is of moderate size, long headed, with a low vault, sloping forehead, a 
high, prominent nose, and wide flaring zygomatic arches. The malars 
are large, and there is, in this respect, a slight mongoloid suggestion, 
One may not, however, on this evidence alone, identify the Royal Clan 
with Turks or Mongols. 

We know very little of the stature of the Scythians. Nine male skeletons 
from the Polish Ukraine, associated with crania of standard Scythian type, 
have a mean of over 170 cm. 65 

It is tempting to find the origin of the Scythians in the previous popula- 
tion of the southern Russian plain. A series of Bronze Age crania from 
the lower Volga region is identical, at least in indices, with the later 
Scythian group, and so is that from the Ukrainian Urnfields. Three 

88 Donicl is of this opinion. He finds the same brachycephaiic type in a collection of 
skulls from an early Moldavian monastery. 

6 Debetz, G., Ann. Lab. Anth. Th. Vovk. Acad. Sc. Ukraine, T. Ill, Kiev, 1930, 
quoted by Doni6i. 

60 Same. 

fil jendyk, R., Kosmos, vol. 55, 1906, sec. 1-2. 

m Ivanovsky, A. (after Donici), TILE, vol. 71, Moscow, 18*91. 

3 Baer, G. E. von, AFA, vol. 10, 1878, pp. 215-231. 

64 Another pronouncedly Corded cranium of Scythian origin was published by 
Majewski, E., in Swiatowit, vol. 9, 1911, pp. 87-88. 

66 Talko-Hryncewicz, J., Przyczynek do poznania, Swiata Kurhanowego Ukrainy. 


skulls of so-called "Cimmerians" likewise show no important deviation, 66 
Furthermore, an important series of Early Iron Age crania from the 
Sevan district of Armenia, probably dated from the earlier half of the 
first millennium B.C., and probably therefore earlier than the Scyths in 
Europe, or at least as early as their first appearance, is exactly like the 
more dolichocephalic element in the Scythian group, and manifestly 
Nordic. The vault, like that of the Scyths, is low, the nose leptorrhine, 
the face leptene, with more compressed zygomata. 67 (See Appendix I, 
col. 38.) Morphologically, these Armenian skulls are characterized by 
a medium forehead slope, moderate browridges and muscular develop- 
ment; a moderately deep nasion depression, and straight or lightly convex 
nasal profile; a projection of the occiput which is most marked in the 
lower segment, and accompanied by some lambdoid flattening; a typical 
compression in the malar region. This series serves a double purpose: to 
show that a Nordic type entered into the modern Armenian blend, and to 
define the Iranian variety of Nordic which may have been likewise in- 
volved in the settlement of Persia and of India. 68 Furthermore, it is very 
similar, both metrically and morphologically, to the early Germanic 
cranial group, and this virtual identity draws together the two geograph- 
ical extremes of an originally united family. 

We have seen that the Scythians and Sarmatians, although they un- 
doubtedly included in their ranks many individuals of different political 
affiliations, formed nevertheless a quite constant principal racial type, 
which was essentially Iranian and a form of Nordic. In its characteristic 
low vault, as in other dimensions, it specifically resembled the earlier 
eastern European and central Asiatic Nordic form. It was essentially a 
member of the racial cluster associated with the spread of Satem Indo- 
European speech in both eastern Europe and Asia. 


We have already dealt with the expansions of two great Indo-European 
peoples, the Kelts and the Scythians, who, during the second half of the 
first millennium before Christ, nearly divided the European continent, 
north of the Alpine mountain barrier, between them. Other groups, such 
as the Thracians, who occupied large expanses of territory in the Balkans, 
have been neglected because of lack of information. 

The first millennium of the Christian era witnessed two more such spread- 
ings of Indo- Europeans; those of the Germans and of the Slavs, the former 

66 Stolyhwo, K., Swiatowit, vol. 6, 1905, pp. 73-80. 

67 Bunak, V. V., RAJ, vol. 17, 1929, pp. 64-87. 

68 Unpublished series of living peoples from the mountainous regions of the northern 
Punjab, and the Northwest Frontier Province, which will be published by Dr. Gordon T. 
Bowles, conform closely to the metrical and morphological specifications of this type. 


to have lasting results in the west, the latter in the east. Unlike the Kelts 
and the Scyths, these two later groups, tardy to receive the civilization of 
the classical world, were destined to people many countries permanently 
with their descendants, and to implant their tongues in many regions. 

Of these two, the Germanic expansion was the earlier. The period of 
Teutonic migration was that of the famous Volkerwanderung, which began 
with the precocious but futile invasion of Italy by the Cimbri and Teutons, 
who fought the Romans between 114 and 102 B.C., and which did not 
end until the adoption of Christianity by the Norwegians in the eleventh 
century put an end to the piratical practices of the Vikings. Its period of 
greatest vitality fell between the second and fifth centuries of the present 

The home of the Germans before their expansion was only in a re- 
stricted sense the modern Germany. The tribes of which this people was 
composed occupied Denmark, southern and central Sweden, Norway, 
and the northern coastal strip of Germany, from the mouth of the Elbe to 
the Baltic shore. The islands of the Baltic near Sweden, namely Gotland 
and Bornholm, were densely populated. 

One must not suppose that these early Germans were the unaltered 
descendants of their Bronze Age predecessors, for there is strong archaeo- 
logical evidence that a new people entered Scandinavia at the beginning 
of the retarded Iron Age of this region. 69 The Hallstatt artefacts are en- 
tirely different in character from those of the Late Bronze Age, and the 
burial rite changed completely, while the old nature worship which the 
Megalithic sea people had brought to Scandinavia now disappeared 
abruptly, being replaced by religious phenomena which we can associate 
definitely with the classical Norse style of worship. The Norse pantheon, 
with its family of gods and its Valhalla, is closely related to the systems of 
Greece and Rome, of India, and of the other Indo-European divisions. 

The principal civilizing agency in the development of the Germanic 
culture was that of the Kelts, but the Kelts were niggardly teachers, for 
they blocked the Germans from direct intercourse with the classical world. 
It was not until the days of the Roman Empire and of the Byzantines that 
the Germans, after driving their way through the vanishing Keltic domain, 
reached these civilizing influences. But the earlier Scandinavians had 
already possessed a distinctive Bronze Age culture, which was not en- 
tirely lost. 

Furthermore, certain strong cultural elements in the time of Germanic 

efflorescence bore strong marks of an eastern inspiration; such as the ship 

burials, which resembled the Royal Scythian interments in every detail 

except for the substitution of ships for wagons; and the art, as expressed 

w Shetelig, H., Falk, H., and Gordon, E. V., Scandinavian Archaeology, pp. 174-175. 


in wood carving, which carried over the richness of the eastern animal 
style, and which reached its highest development in Norway. The Ger- 
mans, like the Kelts, had been subjected to a very strong influence from 
the plains to the east. 

Linguistically, the early Germanic tongues were much in the debt of 
the Kelts. Many of the words needed to express new things were of Keltic 
origin. Hubert, the Keltic authority, believed that the Germanic lan- 
guages were the garbled borrowings of some Indo-European speech by a 
people to whom the Indo-European phonemes were difficult. 70 It is true 
that consonantal shifts from K to H, and the like, are more extreme than 
those in other Indo-European languages. It is very likely that the ances- 
tral Germanic speech was introduced into Scandinavia by the invaders 
who brought the Hallstatt culture to that backward region. 

It is the task of the physical anthropologist to help the archaeologist and 
linguist discover the identity of these Iron Age invaders, whose arrival in 
Scandinavia cannot be put back earlier than the sixth or seventh cen- 
turies B.C. This should be relatively easy, for the newcomers buried while 
the older population presumably continued cremating their dead. The 
Danish series is the most extensive, with 42 adult male crania 71 (see 
Appendix I, col. 39) ; of these only one has a cranial index of over 78. The 
series is strongly dolichocephalic, with a mean of 72.3. There is no trace 
of the brachy cephalic element which had been so important in Denmark 
from the beginning of the Neolithic through the Bronze Age. 

The Danish Iron Age crania form a homogeneous group. They belong 
definitely in the same class with the other Iron Age Nordics of Lausitz 
Urnfields inspiration, and more particularly the purely long-headed ele- 
ment in the Keltic blend, for the low vault and cylindrical transverse pro- 
file of the Keltic crania are also common here. Except for the lesser breadth 
of head and face, and greater vault length, they closely resemble the 
Keltic crania of Gaul and of the British Isles, and those of the Scythians, 
while they are virtually identical with the Armenian Iron Age skulls dis- 
cussed in the last section. The Danish Iron Age crania, then, are probably 
the same as those of the ancestral proto-Kelts before their arrival in south- 
western Germany, and of the ancestors of the Scythians and eastern Ira- 
nians. These Danes were a tall people, however, for the stature of 25 males 
was 171.5 cm. This agrees with that of the earlier peoples of the same re- 
gion, and with that of the Scythians. 

In this Danish series there was, without doubt, a selection on the basis 
of differential methods of disposal of the dead; the numerous Bronze Age 

70 Hubert, H., The Rise of the Celts, pp. 50-52. 

71 Nielsen, H. A., ANOH, II Rakke, vol. 21, 1906, pp. 237-318; ibid., Ill Rakke, 
vol. 5, 1915, pp. 360-365. Reworked. 


population, compounded of Megalithic, Borreby, and Corded elements, 
could not have disappeared completely. After the various elements in 
the Danish population have had time to blend, we shall see them 

The Swedish population of the Iron Age, best represented by a smaller 
group of 14 males 72 (see Appendix I, col, 40), was essentially the same as 
that in Denmark. There are, however, a few differences the vault is 
higher, the face wider, the upper face shorter. Perhaps these more 
peripheral Scandinavians showed a little of the older blood. 

During the Iron Age, Norway was, for the first time, definitely settled 
by people comparable in civilization to those in Denmark and southern 
Sweden; it is likely that many of the earlier inhabitants of Jutland and the 
Danish archipelago had fled to the southwestern corner of that country, 
while other migrations came across from southern and central Sweden. 

The most extensive Iron Age series from Norway is that of Schreiner, 
which contains 27 male crania. 73 (See Appendix I, col. 41 .) These are quite 
different from those of either Denmark or Sweden. They are larger and 
much more rugged, with heavy browridges and strong muscular markings. 
Metrically, they approach the Upper Palaeolithic series of Morant; and 
they could fit easily into the range of the central European Aurignacian 
group. The Mesolithic crania of Stngenas and Mac Arthur's Cave would 
not be out of place here. Yet in most dimensions, they fall a little short 
of the Upper Palaeolithic mean. 

They are purely dolichocephalic, with a cranial index of 71.7. On the 
whole, they are just what one would expect from a Danish Iron Age 
Upper Palaeolithic cross, with the latter in the majority, and this explana- 
tion agrees well with the archaeological data. The stature, 169.5 cm., fits 
both types. There is another possibility, however, that they had a strong 
Corded element. That some Corded blend entered into this mixture was 
indeed likely, but it is impossible to substitute the Corded for the Palaeo- 
lithic element, since the high vault of the former is not in sufficient evi- 
dence, and the faces of the Norwegians are wider than either Corded or 

The central coastal Norwegians of the Iron Age must have been in part 
true descendants of the Upper Palaeolithic people of central Europe, who 
moved northward and westward with the retreat of the last ice, and re- 
mained relatively undisturbed in the centers of its last melting until the 
arrival of new immigrants in the Iron Age. There must, however, have 
been regional differences of type in Norway at this time which persisted 
until the modern period; late Viking Age series from Jaeren, T0nsborg, 

78 Rctzius, G., Crania Suectca, reworked. 

78 Schreiner, K. E., SNVO, II, #11, 1927, pp. 1-32. 


and Skien 74 in the south show the presence of a brachycephalic type, 
massive in build and of great cranial size, which is metrically related to the 
Borreby group of Denmark and northern Germany. These may represent 
colonists or refugees from Denmark. 

A late group from Sogn, 76 in the north, includes mesocephalic crania 
with extremely low vaults and smaller dimensions, associated with black 
or brown hair preserved in the graves. Metrically, they suggest modern 
Lapp crania in most respects, and serve to mark the northern Norse 
borderland, beyond which Norwegian settlements were, in the Viking 
period, only sporadic. These various series place Norway for the first time 
in history in the full light of physical anthropology, and show that the land 
of the Vikings was the last periphery of the Nordic world, in which ancient 
but fully evolved forms of humanity blended with the newcomers from the 
south and east. 

Linguistically, the Germanic peoples who invaded other parts of Europe 
from Scandinavia and North Germany have been divided into two groups: 
East Germans and West Germans. The speakers of East Germanic in- 
cluded the Goths, Vandals, Gepidae, and Burgundians. The Goths 
claimed to have crossed the Baltic from Sweden (not from the island of Got- 
land) to the mouth of the Vistula. The Vandals and the Gepidae presum- 
ably had the same origin. From the Vistula, the East Germans expanded 
southward and eastward into the Scythian country, where the Gepidae 
seized control of Hungary, and the Goths finally established an important 
kingdom on the north shore of the Black Sea. 

From here, the history of these tribes is well known. They all had im- 
portant relationships with the Roman Empire, and adopted Christianity. 
The movements of the Goths into Greece, Italy, and France do not merit 
detailed description. The Visigoths pushed westward, occupied southern 
France shortly after 400 A.D., and moved down into Spain where they were 
gradually absorbed into the population of the northern provinces. The 
eastern Goths who fell under the rule of the Huns met a similar fate. 
Of a once numerous and mobile Gothic nation no trace remains. The 
same is true of the Gepidae, and of the Vandals, who went from eastern 
Europe to France, Spain, and North Africa, whence they were subse- 
quently deported to Byzantium. No doubt, Gothic and Vandal blood 
flaws in the veins of some modern Spaniards as well as of the peoples in 
other countries through which they passed. But this eastern branch of 
the Germans failed to make any lasting impression upon the racial map 
of Europe. 

Although there is not much data concerning the physical type of these 
eastern Germans, there is enough to enable us to come to some definite 

w Larsen, C. F., SNVO, #5, 1901, pp. 3-53. Ibid. 


conclusions. A series of Goths from the Chersonese north of the Black 
Sea, dated between 100 B.C. and 100 A.D., includes three male and eight 
female skeletons. 76 All of these are long headed, and they belong to a 
large, powerful Nordic type which reflects their Swedish origin, for they 
are no different from the Swedish Iron Age crania which we have already 

A later group of Gepidae dated from the fifth or sixth centuries in Hun- 
gary shows the persistence of this same type; despite historical blending 
with the Huns, of eight skulls at our disposal, all but three fail to show def- 
inite traces of mongoloid mixture, and in these three the non-Nordic traits 
are not manifested metrically. One is forced to the conclusion from this 
series, as from that of the Goths in the Chersonese, that the East Ger- 
manic peoples who took part in these wanderings preserved their original 
racial characteristics so long as they retained their political and linguistic 

The same conclusion results when one examines the Visigothic skulls 
from northern Spain which date from the sixth century A.D. 77 Here a 
series combined from several cemeteries shows us exactly the same Nordic 
type, with tall stature and with a high-vaulted skull, a long face, and a 
broad jaw; in this respect resembling, in a sense, the earlier Hallstatt 
crania, but more particularly those of the western Germanic group, es- 
pecially the Hannover Germans and the Anglo-Saxons. 

The western branch of Germanic-speaking peoples, while historically 
less spectacular, was destined to be far more important in the eventual 
peopling of Europe. This included the ancestors of the Anglo-Saxons, 
of the Frisians, and of the Germans proper. Among the latter may be listed 
the Franks, the Alemanni, the Bavarians, the Thuringians, and the 
Chatti, whose descendants are the Hessians. Under the Franks may be 
listed the ancestors of the Flemish- and Dutch-speaking peoples whose 
closely related languages are a mixture of low Franconian and Saxon 
elements. All of these peoples worked their way southward, and in some 
cases westward, gradually and without ostentation; the Alemanni to 
Switzerland and Austria, the Bavarians to the principality which bears 
their name, the Thuringians to Bohemia as well as to Thuringia, and the 
Franks to the upper Rhine country, Belgium, and France. The Bur- 
gundiansj members of the eastern branch of Germans, sophisticated like 
the Goths from contact with the Roman Empire, crossed the Rhine ahead 
of the Franks, and occupied Rhenish Gaul at the same time that the 
Vandals were admitted under Roman sanction. ' 

76 Schliz, A., PZ, vol. 5, 1913, pp. 148-157. 

77 Barras de Aragon, F. de las, MSAE, vol. 6, 1927, pp. 141-186. 
P6rez de Barradas, J,, MSAE, vol. 14, 1935, pp. 141-172. 


The prototype of the western German peoples who migrated from the 
region about the mouth of the Elbe is well represented by a series of skulls 
from Hannover which includes 41 male crania. 78 (See Appendix I, col. 
42.) Metrically, these differ from the Danish Iron Age skulls in being 
slightly longer, somewhat broader, and considerably higher. The fore- 
heads are broader, and the face is wider, and in many cases a bit longer. 
These skulls deviate from the normal Nordic type of central European 
origin with which we are familiar in their greater size and robusticity, and 
particularly in their greater vault height. 

The skulls of the Anglo-Saxons who invaded England in the fourth and 
fifth centuries of the present era 79 (see Appendix I, col. 43) are almost 
identical with this Hannover group. It is to this same specific category 
that the Spanish Visigothic skulls to which we have already referred belong. 
To it must be added two series of old Frisians from northern Holland, 80 
which are identical in every respect. The skulls of these old Saxons, old 
Hanoverians, and old Frisians differ in a number of ways from those of 
other Nordics which we have studied. They are larger than the Aunjetitz 
group and the Danes, and in fact any other series of Indo-European speak- 
ers that we have met, except the Norwegians. They lack the low vault and 
sloping forehead common to the earlier Nordics of Denmark, the Gauls, 
and the Scyths. The vault is moderately high; while the cranial index 
is on the border of dolicho- and mesocephaly. Compared with the other 
Nordics, the forehead is relatively straight, the browridges are greater, 
the muscular markings more pronounced, the cranial base wider, the 
face longer and somewhat wider. 

The type represented by these three groups and by the Visigoths seems 
to be a variant of the Nordic type to which the early Indo-European 
speakers belonged. Its difference is one of size, and it appears to have at- 
tained this distinction through a mixture, in southern Scandinavia and 
Germany, between the older local population, consisting of a combina- 
tion of Megalithic, Corded, and Borreby elements, and the purely Nordic 
Danish Iron Age group. The resultant typfc approaches in some respects, 
but does not even approximate in size, the coastal Norwegian population 
which we have already studied, and it deviates far less from the central 
European Nordic than does the Norwegian group. 

This physical type is accompanied by tall stature, of about 170 cm., 
and by a considerable heaviness and robusticity of the long bones. The 
bodily build was clearly heavier and thicker set than that of the previously 
studied Nordics. That it was characteristically blond is attested by the 

7 * Hauschild, M. W., ZFMA, vol. 25, 1925, pp. 221-242. 
79 Morant, G. M., Biometrika, vol. 18, 1926, pp. 56-98. 
>Reche, O., VUR, vol. 4, 1929, pp. 129-158, 193-215. 


pigmentation of living examples as well as by numerous early descrip- 
tions. This type, being a mixed variety of central European Nordic com- 
bined with old northwestern European elements, is not a true Nordic 
in the sense in which the word has been used in this work, and its common 
and exclusive designation as Nordic in popular parlance as in scientific 
works is responsible for much of the confusion prevalent in the identifica- 
tion of that racial type today. Since it is found among both West and 
East Germans of the period of dispersal, it is essentially the Germanic or 
Teutonic racial type. The eccentric linguistic position of the Germanic 
peoples in the total Indo-European family has its racial connotations. 

One of the principal outlets for this movement from the northwestern 
coasts of Germany was the Anglo-Saxon invasion of the British Isles. 81 
This had begun by 250 A.D., when the Saxons raided the southern and 
eastern coast of England. It was a period of general turmoil, for Irish 
pirates were plundering the coast of Wales at the same time. The Romans 
were hard put to defend themselves against this double peril, and despite 
their military and naval precautions, the raids grew in volume and fre- 

In 406-407 A.D., large invasions of Germanic peoples crossed the Rhine 
and pillaged the Roman settlements in most of Gaul. This broke off com- 
munications between Rome and Britain. With Gaul out of Roman con- 
trol, there could be no hope of holding Britain. Hence, in 409 A.D., the 
Emperor Honorius issued a decree bidding the inhabitants of Britain to 
shift for themselves in the future. From this point on the Saxons received 
little opposition, and settled in great numbers. Since the Saxons were 
not townsmen, they did not occupy the cities which they plundered, and 
the urban population established by the Romans in England maintained 
its identity for a century or longer before the towns were abandoned or 
became Anglicized. 

The earliest Saxon contacts were Viking raids in which they not only 
pillaged the coastal settlements but also rowed far up the rivers, estab- 
lishing temporary camps in the upper waters. When the main body of 
Saxons under Gerdic marched from the region of the Wash across Lincoln- 
shire to the upper Thames Valley, the invaders found that other Saxons 
of more temporary habits had preceded them. Hence it is necessary, in 
studying early Saxon remains, to distinguish between mixed communities 
in which raiders had taken native women to wife, and pure Saxon settle- 
ments in which whole families and villages had emigrated at the beginning 
of the period of serious settlement. 

The Saxons occupied, for the most part, empty country. This was be- 
cause they were accustomed to low-lying land with a deep, rich soil, and 

81 Kendrick, T. D., and Hawkes, G. F. C., Archaeology in England and Wales, 1914-1931. 


had formed, in their earlier home, the habit of tilling this in strips with 
deep ploughs drawn by eight oxen. The Kelts, whose agriculture was 
more cursory in character, preferred the uplands already made treeless 
by nature, and cultivated in square fields. They remained for the most 
part on territory frequented by the Bronze Age and Neolithic men be- 
fore them. The Saxons, who liked forests as well as lowlands, cleared 
the marshes and river valleys of trees, and drained and planted them. Ow- 
ing to this fundamental difference in methods of agriculture, the two peo- 
ples overlapped little at first, and the Saxons and Britons occupied adjoin- 
ing territories in many parts of England for several centuries until at length 
the Saxon social and political domination submerged the language and 
culture of the earlier inhabitants beneath its own pattern. 

The Anglo-Saxon skeletons which have been described earlier are de- 
rived from the graves of the heathen period, from the fifth to the end of 
the ninth centuries. The skulls from these graves 82 make a striking 
contrast to the Keltic Iron Age type which preceded them. While the 
Iron Age forehead is extremely sloping, that of the Anglo-Saxon skulls 
is rather steep and high, and the skulls which possess mandibles show 
that the Anglo-Saxon type was deep jawed, with a great distance from 
lower tooth line to chin and with a long, sloping ascending ramus. The 
cranium as a whole is steep sided with a well-rounded occiput, and fre- 
quently lambdoidally flattened. 83 The browridges are moderate to heavy. 
The nasal bones are highly arched, with often a considerable nasion de- 
pression. Muscularity of a pronounced character is indicated by deep pits 
and ridges on the long bones, which are thick and heavy. Compared with 
the Iron Age people, the Saxons were large bodied, and their more con- 
siderable body weight is correlated with a larger braincase. The mean 
stature of various series of Anglo-Saxons ranges from 167-172 cm. 84 and 
the total mean equals 170 or 171 cm. 

Although there was a difference in the localities from which various 
groups of Anglo-Saxons came, little regional difference is manifest in the 
series from England. The Jutes who settled in Kent, and who came from 
the peninsula of Jutland, seem larger faced than the Saxons themselves, 
but the difference is actually slight. 85 In the total Saxon group studied 

82 Morant, Biometrika, vol. 18, 1926, pp. 56-98. 

Brash, J. C., Layard, D., and Young, M., Biometrika, vol. 27, 1935, pp. 388-408. 

83 Lambdoid flattening is a characteristic common to Neanderthal and Upper 
Palaeolithic man, but rare in the exclusively Mediterranean group. 

84 Calculated from a number of series, involving over 120 adult males. Sources: 
Beddoe, J., JRAI, vol. 19, 1889, pp. 2-11. 

Duckworth, W. L. H., PCAS, vol. 27, 1926, pp. 36-42. 

Hooton, E. A., JRAI, vol. 64, 1915, pp. 92-130. 

Humphreys, Ryland, Barnard, etc., Archaeologia, vol. 73, 1923, pp. 89-116. 

Mortimer, J. R., Man, vol. 9, 1909, pp. 35-36. 8fi Morant, loc. cit. 


by Morant, both males and females belong to the same clearly differen- 
tiated type, and there is no confusion between them and the Iron Age 
form. They thus preserved their racial identity at least until the end of 
the eighth century. 

A number of individual cemeteries, which date from the earliest period 
of Saxon invasion, give us a lively picture of the manner in which the first 
Saxon raiders and settlers operated. One of these is the graveyard at 
East Shefford, Berkshire, containing eight male and twelve female adults, 
as well as eight infantile and juvenile specimens. 86 All of the adult males 
thirty years of age or older represent a single type, the classical Saxon, 
and all are long headed. One of the females belongs to this same type, 
and she was buried differently from the other women, with horse trappings 
in her grave. The rest of the women were rounder headed, with cranial* 
indices going up to 82.4, and some of them were planoccipital. They had 
wider, shorter noses, some prognathism, and shorter, shallower jaws. The 
adolescent women seem to be a blend of these two types. Although many 
of these differences may be due to sex and age, others, such as the funda- 
mental head form, are clearly racial. 

This cemetery presumably represents a raiding party which settled in 
the upper Thames waters before the onset of the mass invasions. It seems 
to have included less than twelve men and only one woman who were 
Saxons. The other women, being Bronze Age descendants, were appar- 
ently British wives of Saxon invaders, while the children were their off- 

The excavation of a round barrow at Dunstable in Bedfordshire throws 
further light on the survival of the Bronze Age physical type into the Saxon 
period. 87 The primary burial of the barrow was a woman of the Early 
Bronze Age; secondary graves contained cremated bodies of the Middle 
Bronze Age, while tertiary burials, heaped in a ditch, consisted of one 
hundred skeletons of persons of the Saxon period who had apparently been 
executed, or slain in battle. One-tenth of them had their hands tied behind 
their backs when they died. Owing to the absence of grave goods, for 
these people were informally slaughtered in a ditch, it is impossible to tell 
exactly who they were. The view that they were Saxon settlers violently 
received by the natives is unsubstantiated. Judging by their racial type, 
they must have been natives slaughtered by the Saxons. 

This series contains a hundred skulls, of which those of 52 males are 
suitable for study. This extensive series resembles the British Bronze Age 
means in most dimensions, but through the narrowing of the cranial vault, 
it indicates a certain degree of mixture with the Iron Age Keltic people. 

* Peake, H., and Hooton, E. A., JRAI, vol. 45, 1915, pp. 92-130. 
^Dingwall, D., and Young, M., Biometrika, vol. 25, 1933, pp. 147-157. 


This excellent series, in agreement with that from Berkshire, proves con- 
clusively that the Bronze Age people did not die out in England but kept 
on mixing steadily with the Keltic invaders and survived racially into 
Saxon times. 

The Saxon invasions of the British Isles were followed by those of the 
Danes, who began raiding the British Isles in the eighth century. The 
Danes, many of whom were actually Norwegians, took the part of England 
in which the Saxons had become densely settled, but they also raided ex- 
tensively in the north of Scotland and in Ireland. Very few skulls of these 
Danes are available for study, but they belong, almost without exception, 
to the expected northwestern Nordic variety. 88 Neither a series of six 
males from the Orkneys, nor of fourteen from various places in Ireland, 
differs from the type of the Saxons. The further Germanic invasion of the 
Normans, after their sojourn in France, took place in such late times that 
the remains of these Normans still repose in Christian cemeteries, and are 
subjected to the same restrictions which protect the skeletons of the solvent 
recently deceased from the hands of the anthropologist. 

The West Germans who invaded Bavaria, southwestern Germany, 
northern Switzerland, and Austria, transformed previously Keltic and 
Illyrian regions into permanent areas of Germanic speech and culture. 
The tribes most fully responsible for this were the Franks, the Alemanni, 
the Bajuvars, and the Thuringians. The skeletons contained in the cem- 
eteries used by these peoples during the first centuries of their settlement 
have been extensively studied, and it is not difficult to determine to what 
extent the Germanic type, as exemplified by the Hanoverians, Anglo- 
Saxons, and Goths was implanted in these regions. 

The Bajuvars, the ancestors of the Bavarians, retained the original 
Germanic head form in their new home, with the cranial index mean of 
75 to 76 in various series. 89 (See Appendix I, col. 44.) Their stature, 
about 168 cm., was moderately tall, and their cranial type, in most if not 
all metrical and morphological features, was reminiscent of their northern 

88 Bryce, T. H., PSAS, vol. 61, 1927, pp. 301-317. 

Martin, C. P., Prehistoric Man in Ireland, pp. 150-151. 

89 Ecker, A., Crania Germanica. 

Henckel, K. O., ZFAE, vol. 77, 3/4, 1925. 
' H6lder, H., AFA, vol. 2, 1867, p. 51. 

Hiis and Rutimeyer, Crania Helvetica. 

Kollman, J., AFA, vol. 13, 1881, p. 215. 

Lehmann-Nitsche, R., BAUB, vol. 11, 1895, pp. 109-296. 

Ried, H. A., BAUB, vol. 16-17, 1907, p. 63. 

Sailer, K., ZFKL, vol. 18, 1934. 

Schicker, J,, MAGW, vol. 35, 1905, pp. 54-55. 

The most satisfactory group is the unpublished series of Mrs. R. S. Wallis of 62 
male and 41 female Bavarian Reihengraber crania measured in the Anthropological 
Institute at Munich. 


ancestors; but in a few of the smaller groups an approximation to the Keltic 
form may be suspected. In every local series, however, the head form 
remains constant, and there are very few brachycephals in any of them. 
The ancestors of the Hessians, if we may judge by a few examples, were 
apparently likewise dolichocephals 90 of the usual North German form. 

The Alemanni may be studied by means of two principal series; a small 
one of twenty skeletons from Oberrotweil in Baden, 91 and a large one of 
over two-hundred from Augst, 92 in the canton of Aargau in Switzerland. 
The series from Baden, while retaining the usual Germanic cranial index, 
assumes in other respects the metrical character of the Keltic peoples 
whom the Alemanni succeeded, and who, as a matter of fact, possessed the 
same cranial index mean of 75 to 76. One must interpret this evidence 
from Baden as an indication that these Germanic invaders were to a large 
extent absorbed by previously settled Kelts, at least in the village which 
used this cemetery and its immediate neighborhood. 

The Alemanni skulls from Switzerland are, as a group, high mesocephals 
with a mean of 78, and include a considerable number of brachycephalic 
crania. On the whole, the total series resembles that of the Keltic prede- 
cessors of the Alemanni, but the stature increased to a mean of 168 cm., 
and the cranial index of the entire group was gradually lowered. In the 
fifth century, 50 per cent of the Aargau Alemanni were brachycephalic, in 
the seventh century, 44 per cent, and in the eighth, 24 per cent. Coinci- 
dentally, the mean cranial index was reduced over this three hundred year 
span from 80.2 to 77.5. Thus the Germanic element, or perhaps a 
Germanic-Keltic blend, increased at the expense of the earlier population, 
and this increase was, as we shall see later, destined to become, in parts of 
Switzerland, permanent. 

The Thuringians, who are known to us through a series from the Saale 
Valley in Germany, and through others from several sites in Bohemia, 93 
practiced the unusual custom, for Germans, of deforming the head by 
annular constriction. Enough undeformed crania are left, however, for 
one to determine their racial type. The Thuringians were purely dolicho- 
cephalic. In none of these groups has a single round-headed skull been 
found. The skulls are, in fact, longer headed than the normal Anglo- 
Saxon and Hanoverian basic type and bear certain resemblances to the 
original Iron Age Danish group, and, at the same time, to the Hallstatt 

90 Virchow, R., ZFE, vol. 9, 1877, pp. 495-504. 

91 Fleury-Cuello, E., ZFMA, vol. 30, 1930, pp. 406-428. 

92 Schwerz, F., AFA, vol. 43, 1917, pp. 270-300. 
9 *Holter, F., JVST, vol. 12, 1925, pp. 1-114. 

Hellich, B., Praehistorickl Lebky v Cechach & Sbirky Musea Krdlovstvi Ceskeho. 
Maly, J., AnthPr, vol. 13, 1935, pp. 37-53. 
Niederle, L., MAGW, vol. 22, 1892, pp. 1-18. 


crania of the same region in which they are found. One may state def- 
initely they are not of Keltic type, and these people had apparently not 
mixed to any extent with the Boii who had preceded them and from whom 
Bohemia derived its name. Like the Boii, however, the Thuringians were 
not destined to remain long on Bohemian soil, for this fertile plain which 
had been subjected to constant farming since the beginning of the Danu- 
bian Neolithic was soon to be taken permanently by the Slavs in the early 
period of their great expansion. 

The Germanic settlement of Austria, including the Tyrol, was a com- 
plicated process, involving the Alemanni, the Bajuvars, the Lombards, 
and the Goths. The Alemanni were the earliest, and the Bajuvars the 
most important. In the mountains, the Lombards settled the southern 
Tyrolese valleys, the Bajuvars those to the north. In the mean- 
while, the Huns contributed a mongoloid element, diluted through 
mixture with the Gepidae. During the seventh century, the picture was 
further complicated by a temporary Slavic expansion which may have 
left human traces in certain of the Tyrolese valleys. Throughout all 
this turmoil, the Romanized Rhaetians still maintained their ethnic in- 
tegrity in the remoter spots, as is witnessed by the survival of Ladino 

A study of the Austrian crania of the centuries of Germanic settlement, 
including for the most part those of Bajuvars, shows them to have been 
largely Nordic, of the usual northern type. 94 A small series of special in- 
terest is that of 26 Lombard crania from two sites: from Nikitsch in the 
OberpullendofT district of Burgenland, and Vinzen, near Regensburg, in 
Lower Austria; both dating from the fifty year interval which the Lom- 
bards spent north of the mountains before their final burst into Italy in 
568 A.D. 96 Eight skulls are those of the usual Germanic variety of Nordics, 
with some exceptionally tall- and large-skulled individuals, while five 
others ranging in cranial index from 77 to 93, show in their flat faces and 
broad nasal bones clear traces of mongoloid mixture. A single male, in the 
Nikitsch series, was strikingly different from the others; a short-statured 
Armenoid or Dinaric, with typical brachycephalic skull, occipital flatten- 
ing, sloping forehead, and other Near Eastern features. He was obviously 
a stranger incorporated into the composite Lombard camp, either a local 
Dinaric or an Asiatic. In earlier times, the Romans had stationed both 

M Geyer, E., MAGW, vol. 61, 1931, pp. 162-194. 
Hell, M., WPZ, vol. 19, 1932, pp. 175-193. 
Merlin, H., MAGW, vol. 16, 1886, pp. 1-7. 
Miiller, G., MAGW, vol. 66, 1936, pp. 345-355. 
Seraczin, A., MAGW, vol. 54, 1929, pp. 323-332. 
Vram, U., RDAR, vol. 9, 1903, pp. 151-159. 
* Miiller, G., loc. cit. 


Syrians and Scotchmen in the Tullnerfeld as garrisons; 96 hence the ethnic 
heterogeneity in this region was chronic. 

The culmination of the overland expansion of the Germans in the south- 
west was the conquest of Gaul by the Franks. Marching from the middle 
and upper Rhineland, they followed the river valleys across Belgium and 
into the valleys of the Seine and Marne, which became the seat of their 
political activities. When they arrived in this region, they were still pagan, 
which was an advantage, for under the leadership of Clovis they were able 
to embrace the currently popular brand of Christianity. This helped 
them to win favor with the Romans, and was an important factor in their 
success. The Gepidae and Vandals, who had become Christian much 
earlier, belonged to the schismatic Arian sect which was then in disfavor. 

These German invaders brought into France and Belgium little which 
was new in the way of material culture, and the continuity of the older 
tradition shows clearly that a racial change in the total population, south 
of the Flemish plain where Frankish is still spoken, could not have been 
complete. During the four centuries of Frankish rule in France and in the 
hilly provinces of Belgium the language of the common people, which 
remained a form of Latin, prevailed over the speech of the conquerors, 
with the result that the national language reemerged as a Romance 
tongue. This sequence of linguistic events stands in striking contrast to 
the situation in England, where Keltic, which had never been completely 
downed by Latin as in France, gave way rapidly and permanently before 
Germanic speech. 

There are enough regional skeletal series of the Frankish period in France 
and Belgium to permit some study of their local characters. The skeletal 
remains from Boulogne 97 and other towns along the English channel are 
all long-headed and of an Anglo-Saxon racial type, which confirms the 
historical record that these regions were settled by seafaring Saxons rather 
than by Franks. The coastal distribution of Saxon place names in Nor- 
mandy and eastern Brittany supports this identification. On the opposite 
frontier of France, at Collognes, near the western end of Lake Geneva, 98 
the descendants of the Burgundians had become brachycephalic, and al- 
most indistinguishable from their Neolithic predecessors who had lived 
at Vaureal, a few kilometers away. 

Aside from these marginal and collateral groups, the Franks themselves 
did not differ greatly from place to place. The most extensive Belgian 
series is that from Cipley in Hainaut, that of France is Mrs. Wallis's 
series drawn from most of the Frankish territory in the northern part of 

9 Lebzelter, V., and Thalmann, G., ZFRK, vol. 1, 1935, pp. 274-288. 
OT Hamy, E. T., Anth, vol. 4, 1893, pp. 513-534; vol. 19, 1908, pp. 47-68. 

98 K>fo Mn ,,.r..;*.. \/r ncAD M A .,^1 a 


the country." (See Appendix I, col. 45.) These series show clearly that the 
Franks were a moderately variable group, but differing as a whole from 
the basic North German type from which they were presumably derived. 
Although individuals belonged to this type, the Franks as a whole re- 
sembled the Keltic peoples who had occupied Belgium and northern 
France before them. This resemblance included the common possession 
of a cranial index of about 76, and a cranial vault height of 132 mm. No 
particular difference can be found between the Merovingian Franks and 
the local Kelts in cranial dimensions or form, except for one important 
fact: instead of falling between the Kelts and the other Germans, in many 
metrical criteria the Franks slightly exceed the Kelts themselves. This 
is true of facial and cranial vault indices. The stature of the Franks, 
furthermore, is on a Gaulish level, with a mean of 166 cm. for males from 
Belgium, and indications that in France it was even lower. 

The conclusion to be drawn from this comparison is that the Franks 
acquired their Keltic-like major physical form in the Rhineland, or the 
southwestern part of Germany in general, before the Saxons drove them 
to France and to the Low Countries. Here, whatever mixture took place 
between them and the previously installed Keltic population made little 
or no racial difference. This conclusion is supported by the evidence from 
Baden, that the Alemanni had likewise, from the beginning of their so- 
journ in southwestern Germany, succumbed to Keltic mixture. Except 
along the Channel coast, the Germanic invasions of France and south- 
eastern Belgium furnished nothing novel to the ultimate racial composition 
of these countries. That of the Kelts, on the other hand, reenforced by 
these Merovingians, was of some importance. 

The summary of our information concerning the racial origins and dis- 
persion of the early Germanic peoples may be stated briefly and simply. 
At the beginning of the local Iron Age, a new people, bearing a Hallstatt 
type of culture, entered northwestern Qermany and Scandinavia. These 
invaders were of the usual central European Nordic type associated in 
earlier centuries with the Illyrians. Through mixture with the local blend 
of Megalithic, Corded, and Borreby elements, these newcomers gave rise 
to a special sub-type of Nordic which was characterized by a larger vault 
and face, a heavier body build, and a skull form on the borderline be- 
tween dolicho- and mesocephaly. 

The Germanic tribes that wandered over Europe during the period of 
migrations belonged essentially to this new type. Exceptions were the 
Alemanni and Franks, who, in southwestern Germany, assumed a Keltic 

w Houz6, E., BSAB, vol. 32, 1 91 3, pp. cix-cxl, for 44 males and 35 females from Cip- 
ley. Mrs. Wallis's series, measured in the Muse*e Broca and the Mus6c d' Historic 
Naturelle, consists of 1 36 males and 66 females. 


physical guise, which they spread to Belgium, France, and Switzerland, 
countries already familiar with the Kelts in person. Other exceptions were 
the coastal Norwegians, to whom for the first time civilization was now 
brought in significant quantity. In the shelter of their chilly fjords the 
new Nordics blended with the hunters and fishermen left over from the 
age of ice, who, through this new genetic vehicle, were assured permanent 


The Slavs, together with their close neighbors and linguistic relatives 
the Baits, stepped relatively late into the theater of European history. 
Speaking an archaic form of the Satem branch of Indo-European, they 
almost miraculously succeeded in maintaining their linguistic integrity 
through the period of obscurity which preceded their time of dispersion, 
despite the widespread activities of the Kelts, the Scythians, and the 
Germans. Slavic is close in many respects to the original form of Indo- 
Iranian, a fact which cannot fail to have cultural and geographical sig- 

It is not yet possible to associate the early, united Slavs with any specific 
archaeological horizon more remote in time than the comparatively recent 
Burgwall moated villages of the early centuries of the present era. Al- 
though all Slavic scholars are not in agreement as to the location of their 
original home, the opinion of Niederle, the dean of Slavic prehistorians, 
bears the greatest weight. 100 He would place it in the densely forested basin 
of the Pripiet River, in northwestern Ukraine and southeastern Poland. 
This region is bounded on the west by the Vistula, on the south by the 
upper course of the Dniester, and on the east by the great forests of the 
former Tchernigov and Poltava Governments. In other words, the Slavic 
ancestors escaped loss of ethnic identity at the hands of the Scythians and 
the Goths through their occupancy of a relatively wooded and swampy 

Their neighbors to the west were Germans and Kelts, who lived on the 
other side of the Vistula; the Baits occupied the side facing the sea after 
which they have been collectively named, while the undivided Finns 
dwelt along the forested stream banks near the sources of the Volga, Oka, 
and Don. The early Iranians, near linguistic relatives of the Slavs, had 
occupied the plains to the south and east, while the Thracians bordered 
the Slavs on the far side of $ie Carpathian mountain chain. 

Like the earliest Iranians and unlike the Scythians, the Slavs were sim- 

100 Niederle, L., ACIA, 2me Session, Prague, 1924, pp. 241-247. For source material 
see his exhaustive series of volumes on the history of the Slavs, SlovanskS Starozitnosti. 
For a recent review of Slavic problems, Sonnabend, H., UEspansione degli Slavi. 


pie farmers and herdsmen. Living in swamps and forests, they had adapted 
themselves to difficult climatic conditions. For some reason still imper- 
fectly understood by the students of population dynamics, they grew in- 
creasingly numerous in the period between the second and fifth centuries 
A.D., and began spilling outward in all possible directions. 

The westward Slavic expansion over much of what is now Germany was 
temporary, for the Germanic peoples themselves soon went through a 
period of eastward expansion during which they Germanized many of the 
new Slavic groups, either by force or by peaceful assimilation. A few 
islands of Slavic speech and culture survived this movement, notably that 
of the Wends in the Saxon Spreewald. The movement of the South Slavs 
took them to the Dinaric mountain chain behind Lower Austria, which 
certain bands crossed to the peninsula of Istria at the head of the Adriatic, 
and into northern Italy itself. The main body moved southeastward along 
the Adriatic coast, following the Dinaric mountain chain to Montenegro, 
and to the Gore region of northeastern Albania. A southern Slavic nu- 
cleus was formed in the kingdom of Old Serbia, centered around Prizren 
and Skoplje. From this nucleus they expanded into the plain of Kossovo 
which, however, they were soon to lose in great part to Turks and Al- 
banians. The Serbs, the most important single people involved in this 
southern expansion, still speak a language closely allied to that of the 
Wends in Germany. 

The movements of the Slavs to the eastward constituted an intensive 
reoccupation of the rich, black earth belt by peasants, for, since Late Neo- 
lithic times, this fertile strip of treeless lowland had been the favorite pas- 
ture and campaigning ground of tribes and nations of warlike nomads, in- 
imical to the full utilization of the ground for tillage. From this black earth 
region the eastern Slavs followed the watercourses of central Russia north- 
ward into the forest country then inhabited by Finns. This upstream 
movement dislodged some of the Finnish tribe^, and brought about their 
historic migration to the Baltic. Many of the Finns, however, stayed 
behind and became Slavicized, mixing with their conquerors. Still others 
remained aloof in small ethnic islands, which even today retain their 
Finnic speech. 

The eastward expansion of the Slavs did not stop with the Urals, but 
gradually continued, after interruptions by Turks and Mongols, into 
Siberia, until finally, in the seventeenth century, its outposts reached the 
Pacific. The Slavs are still growing more numerous and still moving east- 
ward. Their period of efflorescence, the latest of the Indo-European ex- 
pansions, has not yet come to an end. 

Since the Slavs continued the practice of cremation well into the early 
centuries of the present millennium, skeletons from the period of unity are 


non-existent, and those from the early centuries of expansion are not abun- 
dant. However, in this instance, literary evidence antedates the osteolog- 
ical, for numerous descriptions of the early Slavs, assiduously collected by 
Niederle, occur in the writings of Byzantines, Arabs, and Persians. 101 
With only one exception, these make the Slavs tall, spare, and blond or 
ruddy. They were often confused with Germans, and this fact strengthens 
the likelihood that they were predominantly of light pigmentation. Only 
one voice was raised to the contrary, that of a Jew named Ibrahim ben 
Yakub, who, having crossed Bohemia in 965 A.D., remarked that the 
Bohemians were surprisingly dark haired. Niederle interprets this solitary 
dissention as evidence that Ibrahim, accustomed to or expecting blond 
Slavs, was struck by a local enclave which differed from the Slavs as a 
whole. In view of the preponderance of contemporary opinion to the con- 
trary, ben Yakub's dissention must not be given too much weight. 102 

If the evidence of literary sources makes the early Slavs Nordic in 
stature and pigmentation, that of osteology makes them the same in the 
metrical and morphological sense. In brief, all of the earliest Slavic 
skeletal material, dating mostly from the eighth to the eleventh centuries, 
falls, by groups if not as individuals, into one or more of the Nordic cate- 
gories already found to be characteristic of Iron Age Indo-European- 
speaking peoples. 

That from Poland, the eastern half of which was included in the home 
of the Slavic peoples before their period of dispersion, is not very abundant. 
Altogether less than 40 male crania may be assembled, and few of these 
have complete measurements. 103 (See Appendix I, col. 46.) These skulls 
are all predominantly dolichocephalic; the mean cranial index is 73, and 
not a single round-headed example is included. Among these Polish skulls 
are some notably long and large specimens with long, narrow faces. The 
noses of the group, as a whole, are fully leptorrhine. On the whole, the 
ancestral Slavs of Poland f were Nordics, within the range of the, Indo- 
European group; these skulls lean to the longer- and larger-headed Corded 
extreme, and resemble in many respects, the Hannover series, and by 
extension, the Anglo-Saxons. 

Numerous remains of the Slavic expansion into Germany show clearly 
the physical types of the particular invaders concerned in this quarter. 
The most important series is that studied by Asmus, who collected the 

w * Niederle, L., AnthPr, vol. 7, 1929, pp. 62-64; also Sl$vanske Starolitnosti, vol. 1, 
1925, pp. 98 ff. 

102 The passage in question has been translated and retranslated through a number of 
languages. I have been unable to find the Arabic original. 

108 Kopernkki, I., ZWAK, part i, 1883. 

Majewski, ., Swiatowit, vol. 9, 1911, pp. 88-94. 

Rutkowski, L., Swiatowit, vol. 7, 1907, pp. 3-21, 22-38. 


skulls of the ancient Wends of Mecklenburg. 104 (See Appendix I, col, 47.) 
These form a reasonably homogeneous group of high dolichocephals and 
low mesocephals, with a moderate vault height, a low sloping forehead, 
long narrow faces, leptorrhine or mesorrhine noses, high orbits, and a 
strongly built jaw. These Old Wends, rounder headed than the Poles, 
fall very close metrically to the Kelts and to the Scythians. In intermediate 
parts of Germany, particularly in western Prussia and Pomerania, the Old 
Slavic skulls are higher vaulted, and closer in this respect to the Polish 
sub- type. 105 

Those in Bohemia are for the most part the same as the Wend crania in 
Germany, except for one series of Matiegka (see Appendix I, col. 48); 
in this, the vaults are extremely high, nearly reaching early Corded dimen- 
sions. This is true to a minor extent of a small group from Slovakia and of 
individual skulls. 106 Thus, in Bohemia, the Slavs included three sub-types, 
with Hallstatt, Polish, and Keltic analogies. 

The Slavs who invaded Styria between the seventh to ninth centuries 
are basically the same as those in Germany, and fall very close to an older 
Keltic mean. 107 They formed, without question, a mixed group and in- 
cluded in their number a minority of round-headed forms. Some of the 
Slavic crania from Styria, recalling the Polish prototype, are extremely 
large and powerful. We have, unfortunately, no data with which to 
trace the further progress of the southern Slavs into the Dinaric mountain 
stronghold, and thence into Old Serbia and the Kossovo plain. We may, 
however, study a third Slavic movement, that which penetrated Russia. l08 

The skulls of these invaders belong to a generalized Nordic form, with 
a cranial index of 75 to 76, and an intermediate vault height. The Ukrain- 
ian skulls from the eighth to the ninth centuries A.D, do not greatly diverge 
from this general standard, but the early Slavic crania from the Moscow 
region in Russia, dated from the eleventh to twelfth centuries A.D., are, in 
fact, almost purely dolichocephalic, with a mean cranial index of 73.5. 

1( Asmus, R., AFA, vol. 27, 1902, pp. 1-36. 
106 Miiller, W., JVST, vol. 5, 1906, pp. 60-77. 
Reuss, K., JVST, vol. 6, 1907, pp. 93-112. 

Schumann, H., ZFE, vol. 23, 1891, pp. 589-592, 704-708; vol. 26, 1894, pp. 330- 
336; vol. 30, 1898, pp. 93-100. 

Virchow, R., ZFE, vol. 23, 1891, pp. 349-350; vol. 24, 1892, pp. 550-555. 

Cervinka, J. L., and Matiegka, J., AnthPr, vol. 3, 1925, pp. 97-108. 

Jelinek, B., MAGW, vol. 20, 1890, pp. 136-147. 

Matiegka, J., AFA, vol. 25, 1896, pp. 150-154. 

Szombathy, J., MAGW, vol. 52, 1922, p. 20. 

Wankel, H., MAGW, vol. 12, 1882, pp. 123-128. 

1OT Toldt, C., MAGW, vol. 42, 1912, pp. 247-280. 

wsDebetz, G., AntrM, vol. 4, 1930, pp. 93-105. 

Derviz, D., RAJ, vol. 12, 1923, pp. 24-38. 

Stefko, W. H., and Schugaiew, W. S., AFA", vol 50, 1932, pp. 44-55. 


On the whole, the Slavic racial type, as exemplified by skeletal series 
from Poland, Germany, Bohemia, Austria, and Russia, was reasonably 
uniform. In view of its geographical location, the Polish group probably 
represents most nearly the original form, while those who expanded south- 
ward and westward absorbed local Keltic and other Indo-European-speak- 
ing populations. The Slavs, like all the other Indo-European-speaking 
peoples whom we have been able to trace, were originally Nordic, and 
there is no suggestion in their early remains, in the regions studied, of the 
numerically predominant brachycephalic racial increments which today 
are considered typically Slavic. However, the Slavs who migrated to 
southern Hungary, like the Germanic Gepidae before them, mixed with a 
local short-statured, broad-faced, and broad-nosed brachycephalic people, 
who, antedating the historic arrival of the Magyars, were descended from 
the central Asiatic Avars. 109 

Most of the Slavs retained their original dolichocephalic cranial form 
until at the earliest the thirteenth, and the latest the fifteenth, century. 
At that time, those who inhabited Russia and central Europe grew pro- 
gressively brachycephalic, at a rapid but consistent rate. Well-docu- 
mented series from Bohemia and the Moscow government show how this 
change progressed from century to century, so that normal means of 73 
to 75 rose as high as 83 by the nineteenth. Few Slavs were spared this 
change, which was parallel to that which affected the southern Germans 
and other peoples of central and eastern Europe. Although it took place 
in the full light of late mediaeval and modern history, no one fully satis- 
factory explanation has yet been offered. 


It is unnecessary to dwell long upon the conclusions reached in this 
chapter. They may be stated very simply and briefly. 

The predominant peoples of the Iron Age in Europe as well as in central 
Asia, the West-Asiatic highlands, and India were Indo-European speakers. 
For some mysterious reason as yet incompletely understood, various 
branches of this linguistic stock underwent periods of rapid expansion 
during which the human beings who spread these languages migrated in 
many directions and disseminated their physical type as well as their 
speech among other peoples. There had, however, been comparable ex- 
pansions before this. The conquest of the cold brought human beings 
into parts of the world where only Neanderthal rfien and lower animals 
had lived, under equivalent climatic conditions, before them. In the ab- 
sence of competition and in the abundance of game, they were able to 
multiply until they were sufficiently numerous to satisfy the requirements 

i* Szirky, S., and HuszAr, G., MAGW, vol. 63, 1933, pp.* 229-232. 


of their environment. The retreat of the ice and the shifting of belts of 
climate had precipitated other movements which may have taken the 
form of expansions, and the discovery of agriculture and animal hus- 
bandry, of course, gave rise to that expansion which Childe calls the 
Neolithic Revolution. 

The Danubian invasion of central Europe from the east may be con- 
sidered as an isolated wing of this movement, that of the swineherds who 
entered Europe from the southeast, another. In the same way, we may 
consider the migration of the megalith- builders by sea; the wanderings of 
the Bronze Age brachycephals, by land and water; and the rapid move- 
ments of the Corded people across the plains of eastern and central Europe, 
as successive and at the same time parallel expansions. Thus, this business 
of expansions was not initiated by the Indo-European speakers. If we 
knew the languages of the peoples who preceded them, we might in each 
case find parallel linguistic as well as racial circumstances. 

The principal point to this chapter is that the Indo-European languages 
were, at one time, associated with a single, if composite, racial type, and 
that that racial type was an ancestral Nordic. We have determined this 
through a study of the skeletal remains of peoples known to have spoken 
these languages at or near the time of their initial dispersion from their 
several centers. The sub-variety of Nordic concerned in each case varied, 
and the variations usually depended upon mixture with other peoples, 
amalgamated during the process of differentiation and expansion. Never- 
theless, the various brands of Nordic so produced were still very much 

Another result of the investigation pursued in this chapter is the dis- 
covery that the mysterious Urnfields people, who began, toward the end of 
the Bronze Age, to destroy their skeletal evidence and did not cease this 
practice until well into the Iron Age, were probably Nordics. Hence the 
smoke veil has been lifted and we may be reasonably sure of what hap- 
pened. Under this screen, the Nordic-like Early and Middle Bronze Age 
peoples of central and eastern Europe became Iron Age Indo-Europeans; 
no important change of race, then, took place in the focus of Urnfields de- 
velopment, that is, in eastern Germany, Poland, and the Ukraine. It is 
likely that no important change of language occurred there either. 

Since, as we have seen, the Early Bronze Age central Europeans were 
racially a Corded-Danubian blend, a concordance of racial facts with the 
most recent linguistic deductions would make the following proposition 

The Danubians who settled the fertile plains and valleys of eastern and 
central Europe already spoke basic Indo-European; the Finno-Ugrian- 
Caucasic blend which produced this linguistic entity took place before their 


migration westward. The introduction of Altaic words, particularly those 
concerned with the care of the horse, were infused into the previous Indo- 
European linguistic blend at the time of strongest Corded influence in 
central Europe, which produced the Aunjetitz culture. 

This reconstruction helps to support Nehring's conclusion that the 
Danubians were the first speakers of Indo-European languages on Euro- 
pean soil, and that Indo-European may be divided into two chronological 
levels without reference to the Centum-Satem division. If the original 
agricultural and cattle-raising complex was connected with the Danu- 
bians, the horse element with its Altaic linguistic connections would belong 
to the Corded. By this argument, we may construct a reasonably com- 
plete concurrence between the three disciplines: physical anthropology, 
archaeology, and linguistics. 

At this point, a word of caution is needed. We must not carry the asso- 
ciations suggested in this chapter too far, and above all we must not form 
the opinion that the terms Nordic and Indo-European are inseparable. 
Indo-European speakers, from the moment of their initial dispersion, 
began mixing with other peoples, and the specific association between 
language and race found in this instance has by now been largely dis- 
sipated. Furthermore, the Nordic race as we have studied it in Europe was 
formed from the union of two or more widely distributed and essentially 
related racial types. It is quite possible and even likely that similar com- 
binations of the same elements took place elsewhere, and that other Nor- 
dics may have arisen without reference to Indo-European speech. Further- 
more, we must remember that, although most Iron Age Nordic groups of 
which we have literary descriptions were wholly or partially blond, we 
cannot be sure that all prehistoric skeletal material which seems Nordic 
in an osteological sense was associated with blond soft parts; we must 
also remember that the "Nordics" in the living sense have no monopoly 
on blondism. 

Chapter VII 

Speakers of Uralic and Altaic 


In the preceding chapter it has been shown that the Indo-European 
languages were probably formed somewhere on the plain of southern 
Russia or western Turkestan, by a blending of languages spoken by peoples 
in a Neolithic or early Copper Age stage of culture. One of the two lin- 
guistic elements in this blend has been positively identified with Finno- 
Ugrian, which at the same time forms one of the two lateral divisions of 
the Ural Altaic stock, the fundamental unity of which is under question. 1 

The blending of Finno-Ugrian with the B element which produced 
Indo-European languages took place at some time no earlier than the 
last few centuries of the fourth millennium B.C., well after the acquisition 
of agriculture and animal husbandry by western Asiatic peoples, and be- 
fore the adoption of a complete Bronze Age technology by the inhabitants 
of the plains north of the Caucasus and the Iranian plateau. The Finnish 
speakers, who contributed so largely to Indo-European speech at that time, 
must have been residents of the plains at the time of their meeting with 
the bringers of Caucasic speech with which their own language was united. 
At the same time, they must inevitably have contributed to the formation 
of the racial blend with which the resulting Indo-European languages were 
early identified. 

The historic Finno-Ugrians, of whom frequent mention has been made 
in the past, with little elucidation, include in the first branch all of the 
Finnish-speaking tribes of central and northern Russia, the Esthonians, 
and the Baltic Finns, as well as the Lapps, who speak an archaic Finnish 
dialect; in the second, the ancestors of the Magyars, the Bolgars, and the 
Siberian Ostiaks and Voguls. 2 At the time of their first historical mention, 
in the classical period, they seem to have been united in central and north- 
ern Russia. The Finns were centered about the middle course of the Volga, 

1 Professor G. J. Ramstedt of Helsingfors University, an eminent student of Altaic 
languages, has come to the conclusion that the Uralic and Altaic groups of languages 
are not, as was previously thought, demonstrably related, but form two entirely separate 
linguistic stocks. He is supported in this view by Professor Szinnyei of Budapest. 
Private Communication. 

2 See Chapter IX, section 8, for a detailed listing of the living and extinct peoples 
known to have spoken Finno-Ugrian languages. 



and west to the country occupied by the Baits and the Slavs; the Ugri be- 
tween the Volga and the Urals. In the sense that they occupied one unified 
territory from which they later spread, they emulated the behavior of their 
Indo-European-speaking neighbors. Movement to the south was in- 
hibited, in historic times, by the presence of the Scythians and Sarmatians; 
before the rise of these horse-nomads, however, they must at some time 
have been in contact with Caucasic-speaking peoples, who may have in- 
cluded the mysterious pre-Scyths, the Cimmerians, the remnants of whose 
speech have been likened to modern Cherkess. 3 

A Finnish expansion took place in historic time, and during the Chris- 
tian era. It consisted of the following movements: the migration of the 
ancestors of the Baltic Finns to the northwest, largely as a result of Slavic 
and Letto-Lithuanian pressure this took place at the same time as the 
Slavic penetration of Russia; the movement of the Bolgars to Bulgaria, 
during the seventh century, and of the Magyars to Hungary, under Turk- 
ish leadership, during the ninth; the migration of the Ostiaks and Voguls 
across the Urals to the Obi drainage, during the thirteenth. 

Before the time of known Finnish expansion, the Scythian barrier in- 
hibited the use of agriculture as a primary means of subsistence among the 
Finnish tribes located to the north of the nomads. Many of the Finns, 
in fact, lived principally by hunting and fishing along the forested streams 
which formed the headwaters of the Volga, Don, and Dniester. But it is 
unlikely that the Finns in pre-Scythian times had been ignorant of agri- 
culture; those who lived in arable country farmed at least by the time of 

The evidence for the racial composition of the early Finns is scanty, but 
incapable of misinterpretation. One small series of ten skulls dating from 
about the sixth century B.C., contemporaneous with the Early Scythian 
period, has been identified with the ancestors of the Volga Finns at the 
time of their unity. 4 (See Appendix I, col. 49.) These come from the 
cemeteries of Polianki and Maklacheievka, from the former Viatka govern- 
ment in Permian Finn country just south of the present Komi or Zyryenian 
Republic. The graves belonged to the so-called Anan'ino cultural horizon. 
This Anan'ino culture 5 was formed from a combination of influences from 
Siberia, the Caucasus, Scythia, and Scandinavia. It did not end suddenly, 
but passed by a gradual process of evolution into the civilization of the 
historic Volga Finns. Therefore, we may consider these skulls, few as they 
are, to represent the ancestors of the Finns before the beginning of their 
historic expansion. 

* Baschmakoff, A., ZFRK, vol. 4, 1936, pp. 194-199. 
*Debetz, G., ESA, vol. 6, 1931, pp. 96-99. 
&Tallgren, A. M., Real, vol. 1, pp. 164-165. 


This small group of seven male and three female crania is not com- 
pletely homogeneous, but it is nearly so. All of the skulls are European in 
racial type. The faces are a little broader than in most Mediterranean 
groups, but not to an exceptional degree. The noses, with the exception 
of one extremely leptorrhine male, are mesorrhine or chamaerrhine; but so 
are those of many early Danubians. The cranial form is mesocephalic or 
dolichocephalic, with one male reaching the figure of 83; the vault is 
moderately high; the forehead usually straight, the browridges moderate. 

There is nothing new about these crania, and nothing specifically 
mongoloid. They closely resemble another small series of eight male 
skulls from the cemetery of Polom in the same district as the Anan'ino 
cemeteries 6 (see Appendix I, col. 50), dating from the ninth century A.D., 
and known to have been those of Finns of the Permian sub-family. In 
view of the small numbers, no difference can be found which would be 
statistically valid. A third group from the Lower Volga, representing the 
Mordvins of the fourteenth century, is similar to the Anan'ino and Per- 
mian crania, except that it is extremely long headed, with low indices, 
centered about the range from 71 to 73. 

When we make a metrical comparison between the first two groups of 
Finnish skulls and all European series previously studied, we find that they 
fit into the ranks of Iron Age Indo-European speakers without difficulty. 
On the whole, they resemble most nearly the larger-sized members of the 
intermediate group; they also resemble the Scythian crania to a consider- 
able extent, and even more the Minussinsk skulls. They are slightly 
smaller than the Germanic type, but equal to it in vault height and face 
breadth. In nose form and cranial height, they resemble the Neolithic 

News of the racial position of these early Finnish skulls will come as a 
surprise to scholars who see in the Finns a group of mongoloid immigrants 
from Asia. But that they were essentially if not wholly European is, de- 
spite the paucity of Debetz's material, incontestable. Nor can one derive 
these Finns from forest-dwellers of Mesolithic tradition, except perhaps as 
a minor influence. Furthermore, in the early Anan'ino series, recognizable 
Corded peculiarities are to be found in but one male skull out of seven. 
The Finno-Ugrians, therefore, may be tentatively considered to have been, 
in the period before they expanded into their historic seats, Europeans of 
mixed origin, basically Danubian in type, with some brachycephalic ele- 
ment and an extremely long-headed variation as well; the latter is already 
familiar to us in the form of the Corded type; the former is not clearly 
definable, but is European. Its only discernible difference from the others 
in the same series is in a greater breadth of the skull. This broad-headed 

8 Debetz, loc. cit. 


element is completely lacking in the late lower Volga group, of which we 
have only the cranial indices. 

Debetz's discovery that the Finno-Ugrian speakers were originally 
purely European in race, and furthermore, not local Palaeolithic or Meso- 
lithic survivors, is in perfect accord with the present state of linguistic 
knowledge, which makes their form of speech one of two equally weighted 
elements in the basic Indo-European. They not only were, but on logical 
grounds must have been, in the larger sense, Mediterraneans. 

On equally logical grounds, this discovery does not invalidate the hy- 
pothesis that the descendants of Mesolithic hunters and fishers persisted 
until modern times in the forests of the far north, nor that some such sur- 
vivors may not have been absorbed by those tribes of Finns which mi- 
grated even beyond the Permian country to the chilly drainage of the 
Arctic Ocean. This theory is very hard to test, however, for if we review 
the early racial history of the northern forest belt, 7 we find very little 
skeletal data with which to work. What material there is comes almost 
entirely from Latvia, Esthonia, and the Ladoga Lake country, all north 
and west of the historic Finnic center. It includes skulls of Corded type, 
both with and without mixture, and a number of ill-defined crania which 
do not fit into the usual European picture. Many of these latter are 
brachycephalic, some are perhaps, but not certainly, incipiently or par- 
tially mongoloid. 

Unfortunately, the manner in which these skulls have been published 
does not permit a lucid review of their racial position. Similar ones 
appeared sporadically in Late Neolithic and Bronze Age series in Poland 
and on the plains of southern Russia, apparently as intrusions from the 
north, but not in sufficient numbers to alter the prevailing character of the 
population south of the forest from which they, as the osseous headpieces 
of stray woodsmen, had wandered. 

Until almost three centuries after the birth of Christ, therefore, Europe, 
except possibly along the very Arctic rim, had not witnessed the invasion 
of any mongoloid people. Western Asia, from the Bosporus to the Indus, 
and the plains immediately east of the Caspian as well, were equally 
ignorant of them. But with the arrival of the Huns this gap was soon 
filled. . 


In order to discuss the movements of Asiatic peoples into Europe from 
the first inroad of the Huns to the conquests of the Osmanli Turks in the 
sixteenth century, it will be necessary to review briefly the events in cen- 
tral and eastern Asia which preceded and precipitated these incursions. 

7 See pages 125-1 26. 


From the time that the Irano-Aryan ancestors had arrived in Russian 
Turkestan in anticipation of their descent into the hills of northwestern 
India, much of this grassy plain had been the home of those Iranians who 
remained behind while their kinsmen climbed the mountains which would 
take them into India and the Irano-Afghan plateau. These Iranians 
apparently developed, or borrowed, a high degree of adaptation to their 
steppe environment, and especially through the perfection of pastoral 
nomadism with the horse as chief instrument of mobility. They expanded 
through the passes to the eastward, which took them to Kashgaria, and 
there came in contact with the Chinese Empire. On the other side, they 
expanded westward into Europe, where we have already studied them in 
the form of Scythians and Sarmatians. 

To the northwest of the vast Iranian domain, in Mongolia, a number of 
semi-agricultural, semi-pastoral tribes, possessing the sheep, probably also 
cattle, and perhaps wagons, but apparently not the horse, came in early 
times to the attention of the Chinese historians. By 800 B.C. we hear of a 
people called the Hiung-Nu, who gradually grew in importance until they 
came to dominate all of Mongolia. 8 At a fairly late date, set by McGovern 
between 541 and 300 B.C., the Hiung-Nu presumably obtained horses, 
and learned to ride them. They seem to have acquired these animals from 
the Iranians or from Turkish-speaking peoples, along with the whole com- 
plex of horse nomadism. Chinese accounts of the Hiung-Nu later than the 
third century B.C. refer to them as typical plainsmen, strikingly similar in 
many cultural respects to the Scythians. 

The six centuries, more or less, from 400 B.C. to 200 A.D., formed the 
period of greatness of the Hiung-Nu in Mongolia, during which they con- 
stantly harried China, and took possession of Chinese Turkestan. Despite 
their conquest, however, Iranian languages, and the mysterious Tok- 
harian B, persisted in the towns until 800 A.D. or later. At length the 
Chinese took measures to rid themselves of this nuisance, and succeeded in 
defeating the Hiung-Nu so completely that they abandoned their territory 
and disappeared to the westward. 

The last mention of the Hiung-Nu in Chinese sources is about 170 A.D. 
-and, exactly two hundred years later, the Huns appeared on the banks of 
the Don in Russia. McGovern has presented a convincing argument to 
prove that the two were the same people; that their passage across Asia 
took them across a space sterile of historians, between the spheres of 
Chinese and of Byzantine chroniclers. Only one glow of light appears in 
this interim; in 290 A.D. Tigranes the Great of Armenia hired some such 
people as mercenaries. 

8 McGovern, W. M., Early Empires of Central Asia. I am indebted to Dr. McGovern 
for permission to make use of his book before publication. 


The history of the Huns in Europe does not require elaborate treatment. 
Having defeated the Ostrogoths and sent them and their kinsmen scurry- 
ing westward, the Huns moved to the present Hungary, which they made 
their headquarters. From here they sent expeditions to Rome, to Ger- 
many, and to France, where Attila was defeated in the battle of the Cata- 
lonian fields in 451 A.D. After his death two years later, the Huns retired 
to eastern Europe, and many of them united with their relatives the Bol- 
gars, who had settled between the Ugrian and Finnic tribes of the middle 
Volga and Kama rivers, where, under Bolgar leadership, a great state 
arose, which flowered between the eighth and fourteenth centuries. 

In the meantime, the Huns in central Asia raided Mesopotamia, Per- 
sia, Afghanistan, and India; presumably the Turkish penetration of cen- 
tral Siberia dates likewise from the period between 200 and 400 A.D. This 
span of two centuries marks the beginning of the great expansion of 
Turkish-speaking peoples, for the Huns, and their allies and relatives, must 
have spoken various forms of speech related to Turkish, many of which are 
now extinct. 

When we view the Hunnish inroad into Europe in the light of the total 
context of Old World history, it ceases to be a strange inruption of hideous 
and invincible barbarians darting out of nowhere, as it at first appeared 
to the Byzantines and Romans. The Huns were a people who had been 
exposed to a high civilization, that of China; they were cultured if illiter- 
ate, and in every sense the match of the frightened adversaries whom they 
met in Europe. When we examine the details of these invasions, we see 
that it was not one simple inroad, but a series of them in which a 
perplexing confusion of names is involved. Chief of the newcomers, after 
the Huns, were the Avars, who arrived in the sixth century. The Huns 
considered these their kinsmen and equals, and later amalgamated with 
them after the Avars had, in the eighth century, been defeated by Char- 
lemagne and had retreated, some to Hungary and others to the Don 

From the fall of the Huns until the rise of the Mongols some thousand 
years later, the history of central Asia is simply a repetition of the same 
theme; some obscure sub-tribe would become important, win leadership 
over the others, and head new invasions of increasing complexity. The 
history of southern Russia became extremely complicated, for the steppes 
of the Don country served as a terminal point for all but the most serious of 
these movements. 

After the Avars came the Turks, called Tii-Kue, hereditary iron-work- 
ers, who had been an old clan of the Hiung-Nu. They defeated the Avars 
in 546 A.D., and settled about the Caspian Sea; from here they conducted 
their raids and expanded, and gave their name to the whole linguistic 


sub-stock of Altaic which all of them, Huns included, seem to have 
spoken. It is probable that their speech superseded many older allied 

In the guise of Petchenegs and Kumans, in the tenth and eleventh 
centuries new waves of Turks moved across the southern Russian steppes 
as far as the Danube. As Seljuks, the Turks took charge of Asia Minor and 
fought the Crusaders; as Osmanlis, they conquered the Seljuks, withstood 
the Mongol advance, captured Constantinople, and swarmed over the 
Balkans and up to Vienna. But meanwhile, in the thirteenth century, 
other Turks under Mongol leaders, now for the first time called Tatars, 
had covered southeastern Europe ahead of the Osmanlis; and, in the four- 
teenth, hordes of true Mongols had followed, leaving permanent settle- 
ments in the Caucasus, the Kalmuck Steppe, and the Crimea. 

In the fifteen hundreds, the tide commenced to turn in eastern Europe; 
the Muscovites grew powerful, and the Asiatic invaders began to draw 
eastward as the steppes were peopled with Slavs. Under the rule of the 
Turks and Mongols, the older population had not entirely disappeared; 
colonies of Alans persisted until the thirteenth century, and Russian col- 
onies lived under the protection of the Turkish Khazars. In the same 
fashion, the Turks and Mongols did not disappear with the Slavic advance, 
and their colonies in the midst of Slavic territory are still numerous. 

There is an abundance of documents dealing with the invasion of 
Europe by the Huns and by their relatives the Avars. These inroads took 
place shortly after the expansion of the Germanic peoples to the east, and 
formed a primary reason for the failure of the Goths and Vandals to found 
a permanent home in the former Scythian country. They took place, also, 
before the major expansion of the Slavs, who moved eastward in the in- 
terim between the invasion of central Europe by the Huns and the whole- 
sale westward migration of the Magyar ancestors under Arp&d. 

That the Huns came in great numbers cannot be questioned, and that 
they introduced a completely alien racial type onto European soil is vividly 
attested by the accounts of numerous contemporary historians, among 
whom may be mentioned Jordanes, Sidonus, Appolinaris, and Priscus. 
These authors unanimously describe the Huns as being short, broad 
shouldered, thick-set, swarthy, flat-nosed, slit-eyed, nearly beardless, and 
bandy-legged. The Avars are described by some authors as being identical 
with the Huns, but by others as being less horrible of aspect. According 
to that Byzantine wit, Jordanes, the Avars defeated the Iranian-speaking 
Alans, who were the descendants of the Sarmatians, by frightening them 
with their faces and not by valor. 

The careful studies of Bartucz, on whose work this following part is 
almost entirely based, has disclosed, in unquestioned manner, the exact 


racial composition of these invaders. 9 (See Appendix I, col. 51.) Many 
of the Hunnish and Avar cemeteries are very extensive, containing, in all, 
thousands of skulls. In many of these cemeteries, particularly in that of 
Mosonszentjanos, purely mongoloid skeletons have been found, unaccom- 
panied by European followers or European mixture. * 

Bartucz finds two clearly differentiated mongoloid types in these cem- 
eteries. The first, which he designates as type A, is dolicho- to mesocephalic 
with a mean index of 75.5 for the males and 77.0 for the females. These 
skulls are of great length and considerable size. The forehead is very 
narrow, the temples sharply curved, and the zygomatic arches laterally 
bowed. The occiput is narrow and conical at the end. From the side pro- 
file, the forehead appears exceptionally low and slanting. The vertex 
falls well back of bregma, and the profile is curved through the extent of 
its length. In the occipital region the line of neck muscle attachment 
forms a powerful torus. 

The vault of this type is lower than that found in any European group. 
It is, in fact, near the low point for mankind, with a range in height from 
120 to 130 mm. The browridges, accentuated by the extreme slope of the 
forehead, are heavy, but the glabella region is flat, the orbits are rounded, 
and with the lower border often projecting farther forward than the upper. 
The nasal bones are long, narrow, and flat; so that the nasal skeleton some- 
times fails to project in front of the malars. The lower borders of the nasal 
opening are smoothly rounded. The malars are extremely large and 
prominent, the canine fossa completely lacking, and the maxillary sinus, 
which overlies it, is so blown out that the surface of the bone is at this 
point often raised. The dental arch of the palate is U-shaped. The man- 
dible is heavy, but the chin, however, but slightly developed. The whole 
sub-nasal portion of the face is enormous. The stature of this type, calcu- 
lated from the long bones, is 164.4 cm. for the males, 153.1 cm. for the 

Type B is also purely mongoloid, but it is brachycephalic, with a mean 
index of 83 for both sexes. The forehead is also low, but much broader and 
more sharply curved, the occiput is rounded and broad, and the skull 
as a whole is globular, although the vault is still low. The face is broad 
and low, the orbits are lower, the nose less leptorrhine, the malars and 
zygomata less pronouncedly mongoloid, than in the case of type A. The 
nasal bones are shorter, the palate broader and rounder, the chin more 
prominent. This type is characterized by shorter stature; 160.9 cm. for 
the males, and 152.8 cm. for the females. 

9 Bartucz, L., ZFRK, vol. 1, 1935, pp. 225-240; Skythika, vol. 2, 1929, pp. 83-96; 
vol. 4, 1931, pp. 75-90; ESA, vol. 5, 1930, pp. 66-73. 
Krecsmarik, E., Dolgozatok, vol. 3, 1927, pp. 160-166. 
Lcbzeltcr, V., MAGW, vol. 65, 1935, pp. 44-46. 


Thanks to the industrious researches of the modern Russian school of 
physical anthropology, it is not difficult to discover the Asiatic relationships 
of these two types. Type A is found today among the living Tungus, 10 and 
it has likewise a long history in Siberia, for it is found among many 
Siberian peoples, including Palaeasiatics, and it is characteristic of many 
of the Neolithic skulls excavated in the neighborhood of Lake Baikal. 11 
Type B belongs to the Mongol-speaking peoples, and is found in especial 
purity among the Buryats, who represent, culturally and probably racially, 
the Mongols before the time of their expansion. Modern Buryat skulls are 
among the largest in capacity known. 

In most Hunnish and Avar cemeteries, type B is more in evidence than 
type A. Type A, however, predominates in the cemeteries which are 
known to have been used by the Huns, type B in those which belong to 
Avars. The Avar cemeteries contain also, in many cases, intermediate 
types which show that these people had begun to mix with members of the 
white stock, either in central Asia, in Europe, or both, and other cemeteries 
in which the white element is in the majority. The leading classes of the 
Huns and Avars, however, appear to have kept themselves apart, and to 
have preserved their mongoloid racial types pure throughout the centuries 
of their political domination. In the graves which are most richly fur- 
nished, and which show that the occupants were men of power and conse- 
quence, the mongoloid types are unaltered. The two graves of known 
Avar heroes contain skeletons belonging purely to type B. 

Bartucz's identification of type A predominantly with the Huns, and B 
with the Avars, seems valid. That the two intermarried freely is shown by 
the fact that in single graves containing a man and wife, the two are often 
of opposite types. In such cases of differential mating, there is no linkage 
between sex and type, indicating that A and B were socially equal. It is 
very likely that the initial amalgamation of these two types took place in 
Mongolia, and not in Europe. Also, the presence of numerous interme- 
diate forms attests this freedom of intercourse. Individual Hunriish skulls 
found as far afield as Lower Austria and France may be easily identified 
with the crania from Hungary, and belong in known cases to type B. 12 

A further light upon the physical characteristics of the Huns is shown by 
a study of Hunnish head hair, from graves of this period. A sample of it is 
very fine, straight, and jet black. 13 In color and in form, this hair was 
classically mongoloid, but this fineness casts some doubt upon the gen- 
eralization that all mongoloid hair must be coarse, especially since it has 

10 Roguinski, A., RAJ, vol. 23, 1934, pp. 105-126. 
11 Debetz, G., RAJ, vol. 19, 1930, pp. 7-50. 
12 Lebzelter, V., MAGW, vol. 65, 1935, pp. 44-46. 
Zaborowski, S., RA, vol. 24, 1914, pp. 318-320. 
"Greguss, P., Dolgozatok, vol. 7, 1927, p. 232. 


been shown that American Indian hair is very variable in this respect. 

The incontrovertible evidence of the Hungarian graves completely dis- 
pels the theory that the Huns may have been largely European in racial 
type. If the Hiung-Nu were ancestors of the Huns, then the early inhabi- 
tants of Mongolia were definitely mongoloid, and belonged to the two 
important racial elements present there today, the Tungus and the 
Mongol proper. This throws the prehistory of central Asia into a clear 
and logical light. It is exactly what one would expect. 

But it is necessary to discover what was the nature of the European 
racial element amalgamated by the Avars. This may be accomplished by 
studying some of the least mongoloid cemeteries. In that of Jutas 14 (see 
Appendix I, col. 52), only five out of twenty-four skulls show any trace of 
recognizable mongoloid features. The Jutas sample, then, may be used 
for testing. Fourteen male skulls are all below 78 in cranial index, and are 
very similar to one of the Minussinsk regional sub-series; less pronounced 
relationships are present between it and Scythian and Armenian Iron 
Age skulls. The resemblance to Slavic and Germanic skulls, which are 
larger, is less pronounced. It is therefore certain that these non-mongoloid 
Avars belonged to the general Mediterranean racial family, and that 
some, at least, were members of the Nordic Iron Age group; it is more than 
likely that they were for the most part incorporated into the Avar ranks 
in central Asia before corning to Europe. The study of the crania from 
another cemetery, that of Tiszadersz 15 (see Appendix I, col. 53), makes 
this virtually certain. 

McGovern has discovered a number of Chinese references to the 
Hiung-Nu and other Turkish -speaking "barbarians" which describe them 
as hairy, big-nosed, and partially blond. In later times, Genghis Khan 
was supposed to be red-haired and green-eyed. It is therefore likely that 
some of the Asiatic Nordic element found in the Jutas and Tiszadersz 
cemeteries was incorporated by the Avars before they left Mongolia, but, 
on the basis of the evidence from purely mongoloid cemeteries Hke 
Mosonszentjanos, it is unlikely that this influence could have penetrated 
the entire Hunnish and Avar nations. 

At any rate, it is evident from the size and number of the Avar ceme- 
teries that, as Bartucz says, 16 these invaders played an important r61e in 
the peopling not only of Hungary but also of adjacent countries of central 
Europe, for the people whom the Avars brought into the Danube basin 
did not depart with the cessation of Avar rule. 

At the same time the Avars did not uproot the former population, which 

14 Bartucz, L., Skythika, vol. 4, 1931, pp. 75-91. 

15 Lebzelter, V., MAGW, vol. 65, 1935, pp. 44-46. 
Bartucz, L., ZFRK, 1935. 

16 Bartucz, L., ZFRK, vol. 1, 1935, pp. 225-240. 


included Slavs and Germans, among older elements, but made them tax- 
paying vassals. Furthermore, in the days of Attila, the richness of the Huns 
had attracted many craftsmen and adventurers to the royal court, among 
whom were many Italians. Priscus's account makes it very evident 17 that 
Attila's capital contained a very heterogeneous population. 

The great migration to Hungary, that which brought the ancestors of 
the present-day Magyars, took place at the end of the ninth and beginning 
of the tenth century, when the Hungarian national hero Arpad led the 
Magyars into Hungary, where many Slavs had settled in the interim after 
the collapse of Hunnish power. We have already seen (p. 220) that these 
Slavs had partially taken over Hunnish physical traits. By 906 A.D., the 
Magyars were at home in Hungary; in the two centuries which followed, 
they adopted Christianity, and invited settlers of many nationalities, 
including Moslems and Jews, to help them occupy the land. These 
newcomers, along with the pre-Magyar Slavs, formed a tax-paying peas- 

The Magyars were Ugrians from the region between the Volga and the 
Urals, who had been partially Turkicized by the Petchenegs and others, 
but had retained their Finno-Ugrian language, albeit strongly shot with 
Turkish. In this respect, they resembled the ancestral Bulgarians, semi- 
Turkicized Finns, who had, a few decades earlier, crossed the lower 
Danube and settled Bulgaria, implanting themselves on a population of 
Slavs who had themselves been but a short while in occupancy. In 
Bulgaria, the Slavic language seeped through and replaced the Finnish; 
in Hungary, the Ugrian became dominant and the Slavic speech to a 
large extent disappeared. Nevertheless, Slavic culture blended with the 
Ugrian and Turkish, to produce modern Hungarian forms. 

We have no physical remains of the early Finnic invaders of Bulgaria, 
but those of the Ugri of the land-taking period, as the Hungarians call it, 
are adequate. As is to be expected, these ancestral Magyars, led into 
Hungary by Arpad, were only mongoloid to a minor degree. 18 Some of 
the crania which are found in wealthy graves do show definite mongoloid 
characteristics, but the others for the most part lack them. The majority 
of the Magyars were of the same Finnish types expected from our previous 
study of Finns in Russia, while smaller minorities included Binaries or 
Armenoids. 19 

At any rate, it was a very mixed population that lived in Hungary dur- 
ing the early Magyar period. On the whole, throwing all elements to- 
gether, the stature was short and the mean head form rnesocephalic. 

17 Brion, M., Attila, the Scourge of God. 

"Bartucz, L., ZFRK, 1935. 


Gsp4r, J., MAGW, vol. 58, 1928, pp. 129-140. 


Since then, the Hungarians have grown rounder headed, as have Rus- 
sians and southern Germans. 

During all the turmoil of the Magyar and Bolgar migrations, the Ugrians 
who remained in eastern Russia passed relatively unnoticed, but in the 
thirteenth century or thereabouts they, for some reason, probably new 
Turkish .pressure, crossed the Urals en masse, and established themselves 
in the western drainage of the Obi. Here they were divided into two tribes, 
the Voguls, on the immediate slopes of the Urals, and the Ostiaks, in the 
lower courses of the tributaries and along the Obi itself. In their new 
home their culture was modified to stiit a more rigorous environment, 
and only those in the southern Obi drainage, at the time of the Russian 
conquest, still practiced agriculture. 

An adequate series of skulls from the time between this eastward migra- 
tion and the arrival of the Russians about three centuries later shows a 
mixture between the original Finnish type, with which we have already 
acquainted ourselves, and Siberian and central Asiatic mongoloids, of the 
two types already found in the early Hunnish and Avar cemeteries. 20 How 
much of the mongoloid blood was acquired in Europe, and how much 
later in Siberia, cannot be determined. 

In the Hungarian period of settlement we already become aware of the 
presence of a new physical type associated with the Turks, who formed a 
minority in the ranks of the Magyars. When we examine the crania of the 
Petchenegs and Kumans, in both Hungary and Russia 21 we see that this 
new type has become the dominant one among these later Turks to arrive 
in eastern Europe. In it mongoloid features are sometimes present, but in 
abeyance. The skulls are very large, of moderate height, extremely brachy- 
cephalic, and planoccipital. The foreheads are sloping, browridges some- 
times heavy, the faces are very broad, and also very long. The orbits are of 
moderate height. The noses are narrow, and although often low at the 
root, frequently project at the bridge, giving indication of a convex profile 
in the living. 

These Kuman skulls, as best represented by Debetz's series which in- 
cludes fourteen adult males, are much longer and broader than historic 
Armenian skulls, 22 and both longer and broader faced. In height, nose and 
orbit dimensions, and the tendency to occipital flattening, these two groups 
are the same. They are also larger than Alpine skulls from central Europe, 
and far greater in facial dimensions; larger too, than the type B mon- 
goloid crania as represented by a large series of central Asiatic Telengets; 

20 Zaborowski, M., BSAP, ser. 4, vol. 9, 1898, pp. 73-111. 
Ssilinitsch, J. P., AFA, vol. 34, 1903, p. 233, etc. 
21 Bartucz, L., AF vol. 1, 1923, pp. 97-99. 
Debetz, G., AntrM, vol. 3, 1929, pp, 89-95. 
22 Bunak, V. V., Crania Armenica* 


much higher vaulted and broader of forehead than the latter, and even a 
little larger faced. 

Thus, the type under consideration, which has become in many regions 
the characteristic Turkish form, is one which cannot be disposed of by the 
simple expedient of placing it in an Armenoid or Dinaric category. In 
size and proportions of the vault, the closest parallel to these skulls is with 
the British Bronze Age crania; but the resemblance here is far from an 
identify, for the British faces, although equally broad, are much shorter. 
In the same sense, the Turkish skulls are reminiscent of the Palaeolithic 
and Mesolithic brachycephalic types from Europe and North Africa. 

Since we know almost nothing of the early skeletal history of central 
Asia, east of Anau and south of the Minussinsk district, it would be worth- 
less to spend too much time at this point speculating on the immediate 
origin of this type. As with so many other problems, we must defer its seri- 
ous consideration to the section on the living, except to point out that in a 
small series of ten skulls from eastern Russian Turkestan, dated between 
600 and 900 A.D., similar but somewhat smaller vault forms are in evi- 
dence. 23 At the same time, a few isolated Turkish skulls, from central 
Siberia, attributed to from the seventh or eighth centuries A.D., 24 are not 
unlike the Kuman crania. 

After the Huns and Turks came the Mongols, who had been later to 
adopt the horse culture of the Asiatic plains. Their homeland was around 
the southern end of Lake Baikal, and they were hunters and fishermen 
before they became plainsmen. The earliest mention of them in Chinese 
history occurs in the seventh century A.D., at which time they camped in 
the country from Urga northward to the forest edge. They are supposed 
to have sprung from a blue wolf, and from this animal to Genghis Khan 
was a span of but eight generations. 

Their conquest of most of the known world began in the first half of the 
thirteenth century, and ended two generations later with the death of 
Genghis Khan's grandson, Kublai Khan. The Mongols were not numer- 
ous enough to do all of their conquering alone, and incorporated most of 
the central Asiatic Turks into their armies. Hence there arose a perplexing 
welter of Mongolized Turks and Turkicized Mongols, and no doubt of 
Mongolized as well as Turkicized Iranians. We have no skeletal material 
adequate to untangle this snarl, but must rely on Mongol and Buryat 
crania from Mongolia itself to determine their racial type. This was 
simply the type B of the Huns, in a relatively pure form, as found today 
particularly among Buryats. Hence the settlement of the Mongols on the 

' 3 Vishncvsky, B. N., KMV, 1921, #1-2. 

24 Gromov, V. I., ESA, vol. 1, 1926, pp, 94-*99. 

Kazantsev, A. I., RAT, No. l-2 4 1934, pp. 129-133. 


Kalmuck steppe brought the pure, brachycephalic Mongol type to the 
country around the northern shore of the Black Sea, and into the lower 
Volga plains, where whole encampments of normal Mongols may still 
be seen today. 

On the whole, the Mongols proper did not influence the racial composi- 
tion of Europe in the sense that the Turks did. Their influence was 
sporadic in most of the regions which they crossed, and strong only in 
southeastern Russia, and in the isolated colonies still living in the Caucasus. 
Elsewhere it merely served to freshen elements already brought by the 
Huns and Avars. 

Lest this survey of Uralic and Altaic-speaking peoples be incomplete, we 
must mention still another group, the Samoyeds, who live east of the Osti- 
aks in the Obi country, and wander along the Arctic shore of Russia as far 
as the Kola Peninsula, where they meet the Lapps. 

The modern Samoyeds, despite their proximity to the Siberian Ugrians, 
belong for the most part to the central, brachycephalic, mongoloid type; 
Bartucz's B group, the classical Buryat-mongoloid. 25 Except in modern 
times, they have had no influence upon the racial composition of northern 


Before indulging in the speculation which the present study of the 
Uralic- and Altaic-speaking peoples in antiquity inspires, a brief review of 
our present knowledge will be in order. Uralic is a linguistic stock or sub- 
stock which includes Finnic and Ugrian, as well as Samoyedic; Altaic in- 
cludes Mongolian, Turkish, Tungusic, and possibly Korean. 

The Finns and the Ugrians were a united people, in the geographical 
sense, until the arrival of the Slavs from the west, and Huns and Avars 
from the east, forced some of them to migrate, and caused the absorption 
of others. Judging by a series of small samples taken from the heart of their 
forest abode, they were members of the general Nordic sub-group, most 
closely related to the Minussinsk people in Siberia, but showing relation- 
ships likewise with Scythians and peoples of known Indo-European lin- 
guistic affiliation. Thus, since the Finns and Ugrians were not Indo- 
European speakers, there is no reason to suppose that all of the nomads of 
central Asia who belonged to this same racial type were Iranians. The 
Samoyeds, distant linguistic relatives of the FinAo-Ugrians, are not rep- 
resented by early skeletal material, and their racial position in antiquity 
cannot be established. 

Sommier, S., APA, vol. 17, 1887, pp. 71-222. 
Klimek, S., APA, vol. 59, 1929, pp. 13-31. 


Of the known Altaic speakers, three branches, the Tungus, Mongols, 
and the Koreans, were and still are almost purely mongoloid. The fourth 
branch, that of the Turks, is the only one the racial origin of which is in 
question. Today most of the Turks are racially European, but in the old 
days the Huns and Avars, who were intimately concerned with the Turkish 
expansion, were as mongoloid as the others, with both Tungus and 
Buryat-Mongol elements represented. 

We are at this point squarely faced with the problem of the origin of 
the living Finns and Turks, and with that of the rdle played by speakers of 
their linguistic stock or stocks in the formation of European and Asiatic 
peoples. These problems may not be finally solved with the evidence in 
our possession. Yet there is enough material, historical, linguistic, and 
somatological, to make speculation legitimate. 

In the foregoing chapter we have seen that the earliest Indo-European 
languages probably moved westward into central Europe as the speech of 
the Danubian immigrants as early as 3000 B.C. These Danubian farmers 
were racially the relatives or descendants of Anatolian and South Russian 
peoples of a special physical type, a branch of the Mediterranean stock 
to which we have given the name Danubian. This type was reasonably 
homogeneous, but the number of skulls upon which its identification is 
based is slight, and it is possible that a minor increment of longer-headed, 
narrower-nosed Mediterranean forms accompanied it, since the two vari- 
ants seem long to have been associated in South Russia. 

Now since Indo-European speech was a mixture of B, or Caucasic, with 
A, or Finno-Ugrian, and since, as we have seen, the earliest known Finno- 
Ugrians were Nordics with a very strong Danubian tendency, it therefore 
becomes likely that the Danubian farmers owed their racial type to a mix- 
ture of two linguistically different ethnic groups who were physically much 
the same, and both predominantly Danubian. 

If we are correct in identifying the Corded people with the introduction 
of Altaic speech into Europe, then the further identification of the Corded 
racial type with (a) the non-mongoloid modern Turks and (b) the Afgha- 
nian racial type of the Irano- Afghan plateau, makes it seem possible that 
there was, in remote food-producing times, an ancestral bloc of peoples 
living on that plateau who spoke languages ancestral to Altaic, and per- 
haps remotely related to Uralic, Sumerian, or both. Some of the peoples 
who formed that bloc presumably moved northward onto the central 
Asiatic grasslands. This change of scene on the part of these early agricul- 
turalists may have had two effects: the introduction of agriculture into the 
oases of Turkestan and into Mongolia, and the development of pastoral 
nomadism by some of the immigrants, with the subsequent rise of the 
horse culture. 


This step in our speculative structure leads logically to the question of 
the origin of the Turks. Having placed Ural-Altaic-speaking white men, 
of a special Mediterranean type still found in Iran and Afghanistan, in 
Turkestan and Mongolia, 26 it is not difficult to suppose that mongoloid 
peoples, originally hunters, were attracted to the plains from their forests 
and rivers by the advantages of the new economy, and that they assim- 
ilated, in adopting it, those of the white immigrants with whom they were 
in immediate contact. 

In the meanwhile, some of the Altaic-speaking plainsmen, related to the 
ancestors of the Corded people, may have mixed with smaller Mediter- 
raneans such as were found at Anau, to produce Nordics of the type found 
in the Minussinsk kurgans, although it is possible that these Nordics do not 
antedate the arrival of the Iranians. An inruption of relatively unmixed 
Corded invaders from their eastern center, about 2200 B.C., brought the 
Altaic linguistic element noted by Nehring in Indo-European speech into 
central Europe, and produced, by a blending of these Corded invaders 
with European Danubian racial elements, the European Nordics, who, 
during the Late Bronze Age and the Iron Age, spread Indo-European 
speech over a wide area. 

In the middle of the second millennium B.C., during the full Bronze Age, 
one branch of these Indo-European speakers, the Iranians, spread east- 
ward from their home in southern Russia across the country north of the 
Black Sea into Turkestan, and thence some of them went southward into 
Afghanistan and India, bearing with them their original cattle and farm- 
ing culture which they had brought from their earlier home, with a min- 
imum of horse culture elements. 

Other Iranians remained on the plains, and took over the horse nomad- 
ism which the Altaic speakers had already developed. That they mixed 
with Altaic speakers, as the legend of the Scythian youths and Amazon 
maidens would suggest, is probable, owing to their acquisition of a low 
cranial vault and a wide face, eastern Nordic traits which at this time were 
foreign to western Europe. The importance of Altaic god names in what 
is known of the Scythian language would support this contention. These 
Iranians spread the horse culture westward to the Danube and eastward 
to China, and pushed those of their Altaic-speaking predecessors whom 
they had failed to absorb northward and eastward into Siberia and 

In Mongolia, about 400 B.C., the horse culture was taken over com- 

26 This is substantiated by the fact that some of the Neolithic skulls from Lake Baikal 
studied by Debetz are of Mediterranean type, while others resemble those of modern 

Debetz, G., RAJ, vol. 19, 1930, pp. 7-50; AZM, vol. 2, 1932, pp. 26-48. 


pletely by the fully mongoloid Hiung-Nu, as indicated by Chinese his- 
torical documents. The royal and noble families of the Huns and Avars 
remained purely mongoloid, but their followers in their march to Europe 
consisted in large measure of these Altaic-speaking white men who accom- 
panied them. The historic Turks are descended in large measure from 
these Altaic-speaking whites. Some, such as the Kirghiz and the Tatars 
whose ancestors invaded eastern Russia in historic times, are half mon- 
goloid; others, including the Turkomans, the Azerbaijani Turks, and the 
truly Turkish element among the Seljuks and Osmanlis, are fully white, 
since their ancestors had never been subjected to this mixture. A third 
group, represented today by the Uzbegs and Sarts of Russian Turkestan, 
and by the pseudo-Armenoid crania found in late Turkish graves in 
Europe, were a mixture of the old long-headed white strain with central 
Asiatic Alpines, such as the Tajiks, and to a lesser extent with mongoloids. 

Mongols, Turks, and Tungus, living today in the forested northern part 
of Asia, that is in Siberia, are historically recent intruders who, in response 
to their new environment, have partially taken over the culture of Palae- 
asiatic aborigines. Their dispersions may be traced from the Altai Moun- 
tains and Mongolia as a center. Their linguistic relationship with each 
other may be due to varying degrees of acquisition of the speech of the 
nomadic white peoples who brought the horse culture to Mongolia, or to 
an earlier diffusion from whites, bringing agriculture to Mongolia, from 
the same source, or to both. The reindeer-milking complex of the Tungus 
and Samoyeds, and the reindeer riding of the former, are borrowings from 
the central Asiatic horse culture. 

The two most important steps in the foregoing reconstruction are: 
(1) the tentative identification of the Corded people with Altaic speech; 
and (2) the identification of the Corded skeletal type with (a) an element 
in the Nordic racial complex of Europe, (b) the living as well as ancient 
inhabitants of Iran and Afghanistan, and (c) the modern Turkomans, 
Azerbaijani Turks, and the true Turkish strain among living Osmanlis. 
The induction of the Sumerians into this argument is helpful if true, but 
not necessary. Some of the Corded cultural paraphernalia had a Sumerian 
Appearance, but this may have been caused by diffusion alone rather than 
by common ethnic ancestry. 

The foregoing hypothesis, in reference to the origin of the Corded peo- 
ple, of the Turks, of the modern Altaic-speaking mongoloids, and of the 
Sumerians, is pure hypothesis and should not be quoted without the in- 
clusion of a statement that it is offered as speculation only. It is not 
intended to form a part of the serious contribution of the present study to 
white racial history. It is included, however, because in the light of existing 
evidence it seems more likely than any other hypothesis known to the 


author which is of equal scope and which purports to explain the same 

In any case, the question of Uralic and Altaic origins is a part of the 
white racial problem, and it is intimately connected with the history of 
Indo-European languages and of the Nordic race. Of two elements in this 
reconstruction we are reasonably sure; that the ancestors of some of the 
living Turks, including the Turkomans, Azerbaijanis, and Osmanlis, were 
always white men, and that the Corded people were racially related to the 
inhabitants of the Iranian plateau in antiquity. 

Chapter VIII 



At this point we have completed the survey in which, with the help of 
the combined disciplines of osteology, archaeology, history, and linguistic 
science, we have attempted to trace the development of racial entities in 
the territory occupied by the white race, from the earliest human times 
to the Middle Ages, the threshold of the modern period. We are now 
faced with the problem of working with a different body of material 
that furnished by the anthropometry of living peoples. We must further 
attempt to fit this material into the frame furnished us by our study of 
the dead, so that from the combination of the two a complete and orderly 
reconstruction will result. 

While we were dealing with the data gleaned from the measurement and 
observation of bones, the chief difficulty which faced us was the lack of 
adequate samples in most of the periods, regions, and cultural units under 
consideration. On the other hand, while metrical accuracy was by no 
means to be assumed, yet the measurements on the dry skulls and long 
bones were for the most part comparable, and technical difficulty was 
subordinate to the paucity of documents. In dealing with the living ma- 
terial, however, we have vastly larger samples. In some countries, as in 
Norway, Sweden, and Poland, these comprise the entire military age group 
of the nation, and thus cease to be samples in the strict sense, and assume 
the character of total populations. In relatively few regions is it necessary 
to use samples of less than one hundred individuals. 

Our authority has, therefore, increased immensely. We may speak with 
some confidence of the superficial physical composition of most European 
nations. But, at the same time, what we have gained in volume, we have 
to a certain extent lost in accuracy, for the present state of anthropometry 
is partly one of confusion and mistrust in regard to technical methods. 
Despite various attempts in the past and in the present to establish a stand- 
ard corpus of technique, l different schools have arisen in different countries. 
What discrepancies may exist between the work of members of each school 

1 Cf. The Geneva agreement of 1912; the standards established by R. Martin in his 
Lehrbuch der Anthropologie; the present laudable attempt of Miss Miriam Tildesley to 
bring about unification. 



can usually be determined and allowed for; but this is not the root of the 
trouble. The chief difficulty is that much measuring has been done not 
by professional anthropometrists but by amateurs, while some with pro- 
fessional status have not been properly trained. Therefore we cannot be 
sure that such men belong to any school, nor that they follow any stand- 
ard other than their own. The accuracy of existing documents on the 
living is far less than that of skeletal data, and it is not always possible to 
know what techniques have been used. This lack of consistency is often 
an obstacle to mathematical comparison, but not enough of an obstacle 
to render many series wholly useless. We still have a better tool for the 
study of race in the living than we had in the documents of the dead. 

Let us review the more important measurements in which technical 
difficulties most commonly arise. Stature, unfortunately, heads the list. 
One would suppose that the maximum height of the body while standing 
would be a constant dimension and one easy to measure, but neither 
assumption is true. Some investigators allow the subject to be measured 
in his shoes, and then attempt to make a standard subtraction for the heel. 
This is seldom if ever satisfactory. On the other hand barefooted negroes 
with horny soles are raised up several millimeters by their callouses, when 
compared to thin-soled white men standing with their shoes removed. 
Differences in posture, and in degree of conscious stretching, may attain 
the dimensions of centimeters. 

Furthermore, it has been established 2 that the human body, except 
in senility, shrinks as much as 2.5 cm. during a daytime spent either afoot 
or in a chair, the amount depending partly on the degree of and nature of 
the day's activity. It makes some differences, therefore, what time of 
day the investigator habitually chooses for his work. At the same time the 
state of nutrition and of health makes some difference, and one must 
beware of series measured entirely in hospitals. 

For the reasons above outlined, and without doubt for others as well, 
we must not, in studying stature as a statistical criterion of racial value, 
even if our samples are equivalent in age, expect to find accuracy down to 
the millimeter. Therefore the common statistical devices used to check 
the validity of the series on the basis of the sampling process are set at 
too- fine an adjustment in view of the coarseness of the measurement itself, 
and in view of the great variability caused by factors other than sampling 
or racial attributes. What applies to stature applies in varying degree 
to measurements of its segments and of other bodily dimensions; the 
breadths of the shoulder and hips, and the diameters of the chest, are de- 
pendent in some degree on the highly variable amounts of sinew, muscle, 
and fat present at the points of measurement. 
2 Backman, G., FUL, N. F. vol. 29, 1923-24, pp. 255-282. 


In the dimensions of the head and face, most of the difficulties found 
in stature and bodily measurements cease to exist. On the whole, a much 
greater accuracy is not only possible but has been attained. There are 
but two important matters in which serious inaccuracies arise with any 
frequency; these are the measurement of auricular head height and the 
location of nasion. 

The first of these, the measurement of the height of the cranial vault, is 
without doubt the least satisfactory of all common anthropometric tech- 
niques. Although technique #1 5 of Martin 3 is considered standard, not 
all use it, and few do it in the same way. Some investigators use special 
metal head-spanners which measure the height of the vault from the 
middle of the ear hole, others measure from the top of the ear hole; still 
others, following Martin, from tragion. There is also a dispute as to 
whether the height taken should be to the vertex, as stated by Martin, or 
to a point exactly above the ear hole when the head is held in an approx- 
imation to the eye-ear plane. 

As a result of these technical difficulties in taking head height on the 
living, differences of from ten to fifteen millimeters exist between the re- 
sults of different investigators working on identical populations, and re- 
ports embodying these discrepancies are published without comment. 
Since the difference between techniques is as great as the difference be- 
tween extremely disparate racial groups of mankind, head height on the 
living is a useless criterion when employed uncritically. Unless the com- 
piler knows the technical peculiarities and personal equation of each 
investigator whose work he uses, he should leave this material alone. In 
the present work, this ruling immediately excludes from consideration the 
majority of published data on head height. 

The second major difficulty, the location of nasion in the living, while 
not quite as inaccurate, is even more serious, since three important vertical 
diameters of the face, morphological face height, morphological upper face 
height, and nose height, are theoretically limited, at their upper bound- 
ary, by this landmark; and nasion is an extremely hard point to determine. 
Ashley-Montagu, however, has recently devised a method which promises 
to overcome this difficulty in most cases. 4 On adult male whites, luckily, 
there is usually enough ruggedness of facial relief to make this difficulty 
less serious than with mongoloids or negroids. Still technical differences of 
from five to ten millimeters render the works of different investigators 
incomparable, and one must again be sure of the individual equation of 
each investigator, or of the school in which he was trained. Since the facial 

3 Martin, R., Lehrbuch der Anthropologie> vol. 1, pp. 185-186. 

* Ashley-Montagu, M. F., AJPA, vol. 20, 1935, pp. 81-93; vol. 22, 1937, #3, Suppl. 
p. 6. 


and nasal indices depend upon vertical as well as lateral diameters, and 
hence upon nasion, these important racial criteria must be taken with 
great reserve, for the constancy of the lateral diameters serves only to 
exaggerate, in the indices, the differences between the vertical dimensions. 
So much for the most serious metrical difficulties. In measurements on 
the living we see a more bountiful but less accurate counterpart of the 
criteria already familiar to the craniologist. There is another large body 
of data, however, unique in living material; the observations on the soft 
parts, including such features as hair form, hair texture; skin, hair, and 
eye color; the shape of the various component segments of the nose, the 
lips, and the external eye. These are important diagnostic racial charac- 
ters and deserve as careful study as do measurements and indices. But, 
unfortunately, accurate comparisons between the work of different in- 
vestigators is even less possible here than with metrical data, since observa- 
tion is a matter of judgment, and no two men's judgments are the same. 
The use of standard pigment scales in determining hair, skin, and eye 
color has helped enormously, but has not entirely eliminated the diffi- 
culties in the pigmentation field. There is no really adequate eye-color 
scale on the market, although Martin's series of sixteen glass eyes is far 
better than nothing. Von Luschan's skin-color scale does not always 
approximate human shades, and this is especially true with whites. The 
Sailer-Fischer hair-color scale, made from actual human hair, is excel- 
lent, in most respects, but has not yet come into common use; the earlier 
Fischer scale, made of bleached and dyed vicufta hair, is also good. 

Unfortunately, however, the majority of our observational data has 
been collected without reference to scales, and published without accu- 
rate definitions, and it is impossible to tell, in many instances, what color 
or what degree of blondism or pigmentation is implied by a given term. 
Then too, environment and age make great differences in pigmentation; 
the degree of tanning or of uncleanliness in regard to the skin color is 
seldom indicated; eyes often grow lighter with age, and the deposit of fat 
in the cornea, called arms senilis^ which gives a grayish-blue tone to the 
peripheral zone of the iris, is often mistaken for eye blondism. Hair color 
is notoriously transitory, changing, in all but pure brunets and extreme 
blonds, continuously from birth to grayness, baldness, or death. 

Most observations, other than those referring to pigmentation and the 
morphology of the pilous system, are divided into the following categories: 
absent, sub-medium, medium, pronounced. These are frequently ex- 
pressed by the symbols, abs., sm., +> ++ Often ssm. and +++ are 
added for greater refinement. In general, the standard for the + or 
medium category is a roughly estimated and ideal mean or intermediate 
white or European male condition. Thus in nasal tip thickness almost all 


negroes would be ++ r ++ + J in beard development almost all Eski- 
mos would be abs., ssm., or sm. There is a tendency for the observer to 
make the mean condition of the people he is studying + or medium, or 
to be unconsciously influenced by his own facial form. 

Various attempts have been made to standardize these quantitative 
observations, and the most promising is perhaps that of the Moscow 
school, where a series of plaster casts has been made to show standard 
stages of sm., +, and H f- in each of the more commonly studied criteria. 
Still, whatever standards are used, the location of the borderline between 
categories must always be a matter of individual judgment. 

Our first difficulty with the study of race from existing data on living 
populations, whether these data be metrical or observational, is therefore 
one of technical inaccuracy and inconsistency. But it is not the greatest 
difficulty which will be encountered, and it is not insuperable. The care- 
ful compiler can usually discover what are the technical idiosyncracies 
of a given investigator, and if he is familiar with the material as a whole, 
he can usually sense improbable divergences from standard technique. 
The comparison of different samples selected from the same population 
by different investigators often makes a standard adjustment possible. 

Technical inconsistencies and inaccuracies render the study of race on 
the living something less than an exact science, but it remains something 
more than a plaything. The manipulation of metrical data requires ex- 
perience and judgment, and the uncritical use of existing materials on a 
purely statistical basis, no matter how erudite in the mathematical sense, 
can never be more than a sterile exercise. Those who employ experience 
and judgment, and who make a discreet use of the simpler statistical 
methods, may learn much from the handling of the immense body of 
anthropometric data. 


In the introduction to the study of the skeletal material, we made only 
the briefest mention of the statistical methods to be employed in that seg- 
ment of the book. 6 This was done because the numerical size and the 
nature of the cranial samples employed limited the treatment, in most 
cases, to a discussion of individual crania and to a comparison of simple 
means. With the living material however, the use of much larger samples, 
and of non-metrical soft part criteria, will necessitate reference to more 
elaborate methods, and therefore a brief allusion to the better known 
statistical principles and techniques which are commonly employed seems 

Modern physical anthropology, in company with other technical and 

8 Chanter I. DD. 14-15. 


biological disciplines, has entered a stage of increasing dependence upon 
mathematics, and lengthy formulae which involve the use of several alpha- 
bets are currently employed by most physical anthropologists. Although 
there are several schools each of which has assembled a favorite collection of 
symbols, the method as a whole is a product of the English biometric school 
founded by Gal ton and Pearson. Aside from the calculation of means, 
the purposes for which these formulae and numerical techniques are em- 
ployed may be reduced to four, which, expressed in the simplest possible 
form, are as follows: 

(1) To determine the degree of homogeneity or heterogeneity of a given statistical 
sample, in the various criteria measured or observed, and to compare it in these re- 
spects with other samples. 

(2) To determine whether or not two statistical samples may be considered 
random selections from a single population. 

(3) Having found that the two samples represent demonstrable different popula- 
tions, to determine exactly how different, in a metrical sense, they are. 

(4) To determine whether or not a given sample is racially mixed, and if it is, to 
discover its component elements. 

Let us review these four purposes and the techniques by which they are 
accomplished, in as simple and brief a manner as possible. 

(1) To study the relative variability of samples. This is done by means of the 
two constants, standard deviation and coefficient of variation. 6 The former, 
in which the variability of the extremes is emphasized by the quadratic 
treatment, indicates how many unit points the average individual in the 
sample deviates from that mean. When used to compare approximately 
equivalent means within the same criterion, it is a simple and useful 
constant. The coefficient of variation is designed to facilitate comparison 
between criteria in which the metrical values of the means are quite 
different, in order to eliminate the size element. By comparing <r's and 
V's of a given sample with those of a general compilation, such as that of 
Howells, 7 one may gauge the relative variability of the sample, and may 
compare it with other specific samples in this regard. This technique is 

6 o- (Standard Deviation) =* *\ ~r A 2 . V (Coefficient of Variation) = 
P. E. M. (Probable error of the mean) 




P. E. a (Probable error of the standard deviation) = . 

P. E. V. (Probable error of the coefficient of variation) = ' 

P. E. Diff. VP. E. M.i + P. E. M.i 
7 Howells, W. W., HB, vol. 8, 1936, #4, pp. 592-600. 


not by its nature limited to living material, but it may be profitably 
employed with many more published series of the living than of crania. 

(2) To test the statistical independence of two samples. The second purpose 
is, in effect, to tell whether or not two samples may be considered separate 
statistical entities. The technique most commonly employed is to com- 
pare the difference between two means with the probable error * of that 
difference. If the difference is three times or more its probable error, then 
the two samples are considered distinct in the criterion under study. If, 
in a large number of criteria, the two samples are consistently distinct, 
then two separate populations are represented. If, on the other hand, the 
two samples are not distinct, owing to the relative smallness of differences 
compared to their probable errors, then we may make one of the following 
deductions: (a) the two groups represent the same anthropometric popu- 
lation; (b) the two groups are really different, but owing to the small 
numerical size of one or both samples, or to the excessive variability of 
one or both, such a difference cannot be established statistically. 

In order to determine which of these two premises is the more likely, 
the exercise of judgment must inevitably be interpolated. If both samples 
are large and of reasonable variability, the two are probably, in fact, 
alike; if both are very small and the probable errors large, the chances are 
great that the samples are statistically worthless. The chief utility of the 
sampling check, therefore, is to find out whether or not apparent differ- 
ences are really of significance. It is not an automatic proof of identity. 

(3) To measure the anthropometric difference between samples. The third pur- 
pose, to tell how close or how distant two samples are in a metrical sense, 
may be fulfilled in any one of a number of ways. One is merely to compare 
the means, and to compute the differences. Then, for convenience, one 
may pool the differences for separate statistical categories. For example, 
the difference between sample A and sample B in head length may be 
4.35 mm.; in head breadth 7.32 mm.; ip head height 1.09 mm. The aver- 
age difference in three vault diameters is therefore 4.19 mm. The average 
for the same three diameters, between sample A and sample C, on the other 
hand, may be 9.73 mm. Therefore we may say that sample A resembles 
sample B, in the totality of three vault diameters, more than it resembles 
sample C. Similarly one may pool the vault indices, or the head and 
face measurements, or the head and face indices, but one may not average 
measurements and indices together. To do so would be to commit the 
kindergarten fallacy of adding oranges and apples. But there are anthro- 
pologists who have not only done this, but who have also added centi- 
meters and millimeters together as equal units, in pooling body and head 

* See footnote 6 on preceding page. 


It has long been the wish of many anthropologists to find some means 
whereby they might express the degree of similarity of difference between 
two populations by a single figure. Taking population A as zero, B 
would be, say, 5.6; C = 7.3; D = 11.9. Thus the relationships of B, C, and 
D in respect to A could be determined. Taking each of the others in turn, 
it would be possible to triangulate and to plot the mutual relationships of 
any number of populations in a simple, graphic manner. Morant, work- 
ing with a formula invented by Pearson, has proposed and employed such 
a method in the form of the coefficient of racial likeness. 8 Some have 
accepted this in principle, others have rejected it. 9 Whatever its theoretical 
validity or error, however, it does actually give approximately the same 
results as a simple pooling of the several categories of differences. Unfor- 
tunately neither a simple pooling nor the coefficient of racial likeness 
takes into account correlative influences which compel several characters 
to vary in concert, and thus to weight, in a variable degree, the totality of 
characters chosen. According to Morant, these correlative influences 
could be eliminated, but only by an unfeasible amount of statistical labor. 

Before proceeding to the fourth purpose, let us pause to make a few 
reflections upon the uses to which the three systems already outlined may 
be put. Although all are useful, not one automatically answers any im- 
portant questions. The first technique, that which is concerned with 
variability, tells us how variable samples are, but not why they are vari- 
able. Unusual variability may indicate an active evolutionary tendency, 
the recent and as yet not fully amalgamated mixture between two popula- 
tions, or any one of a number of other causes. Unusual homogeneity, on 
the other hand, does not necessarily mean racial " purity," in the historical 
sense, but rather a complete amalgamation and a static evolutionary 
condition. The second is useful mainly to eliminate from serious consider- 
ation statistically inadequate samples. The third gives a detailed idea 
of degrees of metrical similarity and difference. But neither the second 
technique nor the third tells the investigator what is the genetic relationship 
between two samples. 

(4) To analyze a racially mixed sample. Let us now turn to the fourth and 
last important use which the physical anthropologist makes of statistics. 
This is his attempt to divide a given sample, which he considers to have 
resulted from a mixture of races, into its component elements, and to see 
what these elements are and how much there is of each in the mixture. 
This is a rather complicated process, and many different methods have 
been devised to accomplish it. 

8 Morant, G. M., Biometrika, vol. 14, 1923, pp. 193-264; vol. 16, 1924, pp. 1-105. 
Pearson, K., Biometrika, vol. 18, 1926, pp. 105-117. 

9 Fisher, R. A., JRAI, vol. 66, 1936, pp. 57-63. 


One is the system employed by Hooton and his school, in which the 
author was trained. That is for the anthropometrist, working either with 
crania or with the living, to divide his series into what seem to him natural 
groupings, and to specify on each measurement blank which of these 
types is represented. After the sample has been seriated as a whole, the 
sub-samples of the different types are seriated separately, and statistically 
compared with each other and with the total mean. By this means it may 
be determined whether or not statistically different elements have actu- 
ally been isolated. If so, the next step is to determine, by comparison, 
what the larger racial relationships of these elements are. 10 Hooton bases 
his system on the principle that the individual possesses a racial identity, 
as well as does the group to which he belongs. 

Another method which is less subjective but wholly arbitrary is that of 
Czekanowski, who plots the mean differences between individuals in a 
sample on a chequered field ; this is done only with indices of the head and 
face, when the original system is followed. 11 Two individuals alike in all 
indices chosen produce a black square at the point where their lines inter- 
sect; two which are less alike produce a square which is striped, in varying 
degrees arranged to show the degree of similarity; then those which are 
dissimilar in all indices are represented by white squares. After these 
squares have been completely plotted, the graph is rearranged so that 
those which are naturally related are placed in contiguous positions. In 
this way it is possible to see how many sub-groups of naturally correlated 
individuals occur, and how large these sub-groups are. The next step is 
to find the racial affinities of each sub-group. For this purpose the Polish 
school has designated a formal list of races, each symbolized by a separate 
Greek letter, and each equipped with a list of ideal metrical positions in 
the more commonly used measurements and indices, as well as with a 
characteristic pigmental position. Each group of correlated black squares 
in the graph is assigned to one of these races, or to a combination of two 
or more, and the percentages of each race in the sample is thus worked 

A third method is that originated by von Eickstedt, the leader of the 
Breslau School, and amplified by Schwidetzky. 12 This method is to sort 
the sample directly into sub-series by splitting the distribution frequen- 
cies of the characters at arbitrary racial boundaries, and by combining 
the results of this process as applied to pairs of characters; to plot the 

10 Hooton, E. A., The Ancient Inhabitants of the Canary Islands; Indians of Pecos; Science, 
vol. 63, 1926, p. 75. 

11 Gzekanowski, J., MAGW, vol. 42, 1912, pp. 17-217; AASF, ser. A., vol. 25, #2, 
Helsinki, 1925; AAnz, vol. 5, 1928, pp. 335-359. 

Eickstedt, E. von, ZFRK, vol. 2, 1935, pp. 1-32. 

Schwidetzky, I., ZFRK, vol. 2, 1935, pp. 32-40; vol. 3, 1936, pp. 46-55. 


distribution curves of the sub-series, so-created, for measurements, indices, 
and percentages of observations; and to test the sorting by a comparison 
of these curves with others which represent arbitrary racial norms. Like 
all such systems, this one operates on the assumption that the result of 

A + B 
mixing A + B in any metrical character is 

The three methods outlined above are all based on the principle of 
correlation. Correlation statistics alone are even more commonly used 
than any of these. One may correlate metrical characters with each other; 
metrical characters with indices; either metrical characters or indices with 
observations, and observations with each other. By means of these corre- 
lation statistics one finds which characters are associated, in the sense that 
their variations are not mutually independent. One finds, for example, 
that light eyes are usually if not always correlated with light hair. The 
elements of blondism are to a certain extent linked. One will also find 
that segments of a dimension are positively correlated with that dimen- 
sion, but this is of no racial significance. If they are not correlated, or are 
negatively correlated, then there is something to investigate. One must 
furthermore expect all gross size diameters to be intercorrelated to some 
extent in any population, for obvious reasons. 

Correlations of racial significance are those which are not dependent on 
gross size and are not involved in a part-and-whole relationship. Thus, 
if tall stature goes with blond hair and short stature with dark hair, or if a 
broad nose goes with a low relative sitting height, and vice versa, then 
the anthropologist who is analyzing his series assumes that he has uncov- 
ered linkages showing racial variations within his sample. 

There is no possible objection to the use of correlations, but there are 
many objections to the ways in which they are often interpreted. In the 
first place, a valid correlation implies some degree of genetic linkage. 
But it does not necessarily imply that this linkage represents with com- 
plete fidelity a combination found in one component element in a hypo- 
thetical mixture. There may have been no mixture at all the group may 
be evolving, by mutation, in a certain direction which involves more 
than a single character. Or if there has been mixture, the correlation 
may represent a recombination of characters. 

Correlation, in brief, shows linkage, but what does linkage mean? We 
must not forget that a population, in the physical as well as in the social 
sense, has an existence of its own in addition to and above the existences 
of its component units, and we must not, furthermore, anticipate the 
findings of the geneticists. All of the methods which partition a series, or 
which employ the principle of correlation, have some justification in their 
initial steps, and some utility, but all of them become unscientific as soon 


as general biological principles are forgotten and arbitrary assumptions 
are allowed to creep into the process of interpretation. 

At this point we must repeat the premise upon which the whole tech- 
nique of the present book is based: The study of race in man is as much a 
social and historical as a biological discipline. Out of his environment man 
creates his cultural milieu, and his cultural milieu, as we are becoming 
increasingly aware, 13 alters his physical nature. When we shall have dis- 
covered some of the laws which govern human inheritance and human 
change, we may become much more mathematical than is fitting at 
present. Laws in biology and in its sub-division sociology, when once 
understood, are seen to be as invariable and as valid as laws in physics. 
But we cannot, and we should not attempt it, remove the study of human 
racial taxonomy from the dimensions of cultural milieu and of history. 
We may and must employ a statistical method, but let it be one tempered 
with simplicity and discrimination, since mathematics to us is not an 
end but a tool. 


(a) Stature and Bodily Form 

Before venturing to draw up a schematic classification of races within 
the white family, let us review some of the better known racial criteria 
from the standpoint of spatial distribution. The use of maps to show the 
distribution of means in a single metrical character is one of the oldest 
and commonest illustrative devices employed in the study of race. It 
has, in fact, formed the basis for several systems of racial classification, 
based upon geographical correlations between two or more characters. 
Such classifications ignore individual linkages in the characters involved, 
and subordinate the position pf the individual as a racial entity. They are 
of necessity based on few characters, and the races so postulated are cor- 
respondingly ill defined. 14 This abuse of cartography should not, how- 
ever, hinder the use of maps in a purely demonstrative sense. 

In this and the two following sections, we present four such maps, 
representing the distributions of stature, the cephalic index, head size, 
and hair and eye pigmentation. 15 These four characters were chosen from 
the total body of criteria because they are the only ones in which it is pos- 
sible to overcome, to a satisfactory degree, the obstacles of paucity of data 

13 Cf. the title and sense of Childe's book, Man Makes Himself. 

14 See Chapter I, section 3. 

16 Attention is called to the earlier maps of Deniker, and of Struck, both of which 
have been extensively copied. 

Deniker, J.,JRAI, vol. 34, NS 7, 1904, pp. 181-206. 

Giinther, H., Rassenkunde des deutschen Volkes, pp. 216-217. (Early reproduction of 
Struck's maps.) 










and technical inconsistency. Not one of the four is completely accurate, 
but all are accurate enough for present purposes. 

In the first of these, the stature map (Map 5), as in those which fol- 
low, boundaries between provinces, linguistic areas, and other ethnic or 
political units have been simplified, and schematized to such an extent 
that the smallest spatial unit recognized is one capable of legible stip- 
pling. Nomadic territories in North Africa, southwestern Asia, and the 
far north, have received an even more schematic treatment. 

The first impression which one receives while examining this map is 
that there seems no orderly scheme; that, except for the stunted circum- 
polar belt, there seem to be no widespread zones of stature. A relatively 
large and consistent area of tall stature, however, is comprised by the 
Scandinavian Peninsula, most of the land area of the British Isles, the 
Netherlands, Finland, the Baltic states, and parts of northern Germany. 
This northwestern European center of tallness is commonly referred to 
in anthropological literature as the primary Nordic racial zone. 16 It is 
difficult, however, to agree that the tall stature of these countries is largely 
the result of the presence of Nordics, since its existence seems to be due 
to multiple factors. Historically, this is precisely the region of maximum 
survival of tall Palaeolithic hunters, while Corded people were concen- 
trated in certain sections of it, especially in Denmark and Esthonia. 
Furthermore other contributing racial elements, such as the Bell Beaker 
people and the Megalithic navigators, were all tall, and these lands under 
consideration are at the same time precisely the regions of Europe least in- 
fluenced by Danubian or Western Mediterranean agricultural invaders. 
Essentially, therefore, these are regions in which all contributing racial 
elements in the past have been tall, and in which there is no short-statured 
ethnic sub-stratum. Furthermore, northwestern Europe has been the 
scene of maximum stature increase during the last century. 

A second European area of tall stature is the Dinaric mountain zone, 
the nucleus of which stretches along a narrow belt from Croatia to the 
Drin River in Albania, and which reaches its peak in Montenegro. Here 
one finds statures as tall as those in the north, and, in the heart of the area, 
taller. The origin of this Dinaric giantism is obscure, since the prehistoric 
archaeology of this region is almost unknown, and the crania documents 
from later times inadequate. We know that the Bell Beaker people set- 
tled here in some numbers, but hesitate to attribute to them alone the 
excessive height of modern Dinarics. 

A third area is found in southwestern Russia, on the northern shore of 
the Black Sea, in the Ukraine; here Atlanto-Mediterranean factors seem 

16 De Geer, S., "The Kernel Area of the Nordic Race within Northern Europe," 
in Lundborg, H., and Hinders, F., Racial Character of the Swedish Nation. 


largely responsible. On Asiatic territory the countries occupied by the 
non-mongoloid Turkomans and by the Iranian-speaking Kurds are seats 
of tall stature, as is the kingdom of Iraq, whose inhabitants have been tall 
since the days of the Sumerians. 

One other principal area of tall stature, which is merely suggested 
within the limitations of the present map, is the Hamitic center located in 
East Africa. One recalls the giantism of the pluvial inhabitants of Kenya, 
which has apparently been perpetuated in the great height of living Ham- 
ites who inhabit the Horn of Africa and the western shore of the Red Sea. 
The most thoroughly Hamitic of the North African Berbers, the Tuareg, 
are as tall as northwestern Europeans. The tall stature zone of northern 
Africa is centered in regions of the Sahara occupied by nomadic Berbers, 
and extends itself into the fertile stretch of Africa Minor where these 
people have settled after invasions. 

Turning to the consideration of short stature, we find that, aside from 
the far north and the territories occupied by recent Mongol invaders, it is 
concentrated today in the very regions most affected by early Neolithic 
migrations of short, food-producing Mediterraneans namely, the western 
Mediterranean countries, from central France to Sicily, and the Danubian 
culture area, especially in its eastern and trans-Carpathian segment. 

In general, one cannot over-simplify a distribution map dealing with a 
character as complex as stature, since south of the Arctic circle there are 
no large zones or major trends, and in most of the sub-areas a complicated 
sequence of historical events has taken place which has brought in a suc- 
cession of peoples with different statures. Furthermore, different environ- 
mental stimuli operating in various places and at varying times have 
further served to complicate the picture. 

The distributions of weight and bodily form, if these criteria could also 
be completely plotted, would make maps as interesting as that of stature. 
What information we possess suggests that they would be much simpler 
and more easily interpreted. In weight, for example, there would be one 
large zone in which the adult males in middle life would average over 
150 pounds, with individuals in the two hundred class common, and this 
zone would correspond to the northwestern area of tall stature, and to 
adjacent parts of Germany, Holland, and Belgium. The center of the 
Dinaric zone would likewise be one of heavy weights, but the rest of Europe 
would run, for the most part, at least twenty pounds lighter. 

In the long stretch of arid countries reaching across North Africa and 
Egypt into Arabia, Iraq, Iran, and Afghanistan, light weights would be 
the rule, regardless of stature, and this would likewise be a zone of pre- 
dominantly linear, or long and narrow, bodily habitus. Stocky build, on 
the other hand, would also be found to have little relationship to stature, 


since some of the tallest northern peoples and some of the Binaries would 
be plotted as lateral. Sex differences in both stature and gross size would 
be found greatest in northwestern Europe, as among Upper Palaeolithic 
peoples, and least in eastern Europe and among western Mediterraneans. 
In general total bulk, regardless of stature, seems partly a function of 
environment, and excessive bodily volume is suggestively centered in 
cool, damp regions, while thin, light-bodied people are most frequently 
encountered in deserts. Great differences in size between the sexes seem 
commoner among large than among small peoples, and are most pro- 
nounced in the regions where Upper Palaeolithic strains survive in most 
concentrated solution. 


(b) Head Form, Head Size, and Other Metrical Characters of the Head 
and Face 

Next to stature, which is of interest to many others besides anthropol- 
ogists, our data are fullest on the cephalic index, for this ratio has been 
the favorite of both professional and amateur students of race ever since 
its invention by Retzius in 1842. The same remarks on the method of 
plotting the stature map apply to that of the cephalic index (Map 6). 
Here the only region of comparative uncertainty lies in the southeastern 
corner, in Iran, where some rather extensive boundary stretching has 
been practiced. 

The distribution of the cephalic index within the area covered by this 
map is a complex affair, and cannot be interpreted hastily. Many factors 
and many events have contributed to this state of complexity, which the 
map only partly represents. One must remember that, as in the stature 
map, the scattered bands and villages of Lapps have been schematically 
united into a nucleus in northern Scandinavia, Finland, and the Kola 
Peninsula. Furthermore minority groups such as Jews, Gypsies, and 
others, have been omitted, since in no region large enough for schematic 
representation are they found in a majority. 

The most striking feature of the map, and in fact, almost its only uni- 
formity, is the steady band of almost pure dolichocephaly which extends 
south of the Mediterranean from the Atlantic coast of Morocco across 
North Africa, Egypt, Arabia, and Persia into Afghanistan; to continue, 
off the map, over Khyber Pass and into the Indus Valley. This band 
represents the greater Mediterranean race in its post-Pleistocene home- 
land. Small spots of mesocephaly in the Moroccan mountains, in Kabylia 
and in the Aures, and along the Tunisian coast, show the relatively re- 
stricted zones of survival of earlier Mediterranean mesocephals and, to a 
lesser extent, of Pleistocene North African men; except for the Tunisian 


coastal centers, where the strong concentration of Punic and European 
populations in pre-Arab times is no doubt partly responsible. 

The extreme long heads, concentrated in the Hoggar and in parts of the 
Algerian plateau, are the Tuareg and the purer families of ancestral 
nomadic Berbers, preserving the head form which they brought from 
East Africa, their Hamitic homeland. The heavily dotted stipple repre- 
sents Mediterraneans of Neolithic age and Arabs, with an infusion of the 
Hamites, while the light dotting represents more clearly the Hamites 
themselves. This is a distinction which should not be pressed too far, but 
which may still be made, for the lightest stippling is found in nomadic 
Berber strongholds. 

Farther east the desert tribes of Libya, and the oasis people of Siwa, are 
extremely long-headed, in a truly Hamitic fashion; the inhabitants of 
Sinai, and some of the tribes in the Nejd, as some of the Mesopotamian 
Bedawin, and groups in Iran, fall into the same category. Here in the 
east we approach the zone of hooked-nosed long heads, quite different 
in facial form from the Hamitic increment farther west. Around the 
Persian Gulf is a ring of higher indices, representing a maritime popula- 
tion which we shall encounter later in the coastlands of southern Arabia, 
off the present map. The long headedness of inland Arabs, whether 
nomadic or agricultural, continues without a break south of the present 
map into Yemen and to the northern and western borders of the Ruba* el 

In Europe itself, long-headed total populations are rare. Only in parts 
of Portugal, in fact, are regional indices under 76 to be found at all. 
Europe on the whole is a brachycephalic or mesocephalic continent. 
Mean indices between 76 and 79, belonging to high dolichocephals and 
low mesocephals with brachycephals in the minority, are found in a few 
places. One, the most continuous area, lies in the northwest; it includes 
the British Isles, most of Holland, parts of Belgium, and the Palatinate 
old Prankish country and most of 'the Scandinavian Peninsula, along 
with the coastal lands of Finland, and with Esthonia and Latvia. 

The regions just enumerated may be considered in a way a unit; most 
authorities would call this, as with stature, the Nordic racial territory, 
and so it is in the accepted sense. Another belt is that of the Iberian Penin- 
sula, the Dordogne Valley in France, Sardinia, Corsica, the Balearics, 
the toe of Italy, and Crete. To this may perhaps be added part of the 
corresponding area in the British Isles, and parts of the eastern site of the 
Balkan Peninsula. This is what remains of the brunet Mediterranean race 
per se in Europe; isolated island groups, a peninsula which throughout 
history has been more African than European, and remnants of the old 
Mediterranean bloc of the shores of the Black Sea and the Aegean. 















Where, we ask, are the descendants of the Danubians, the Aunjetitz 
Nordics, and their Iron Age successors in eastern and central Europe? 
Only in the mesocephalic belt across eastern central Russia, and the region 
immediately north of the Caucasus, and again in the central and eastern 
Balkans, do traces of the original head form of these peoples appear, 
emerging as that of a population bound to the soil. Perhaps in the tall 
stature and high mesocephaly of the Don country there is also some trace 
of the Scythians. The country between the northern shores of the Caspian 
and the middle Baltic does indeed form a zone of relative long headedness 
between the mongoloid brachycephaly of central Asia and the European 
brachycephaly of central Europe. 

This central European brachycephaly may not be treated as a com- 
pletely unified entity. In the first place, we find its westernmost nucleus 
in southern France in the Massif Central, which is the home of the Alpine 
race in its truest form. Here extreme round headedness such as is seldom 
exceeded elsewhere in the world is located. The valley of the Rhdne forms 
a partial gap, beyond which lies another brachycephalic zone in eastern 
France, especially in Burgundy and the Jura, and adjacent portions of 
Belgium. Here again we find a high zone of brachycephaly, accompanied, 
as we have seen, with a greater stature than that found in the western 
Alps, and as we shall see later, a lighter pigmentation. Here is another 
brachycephalic nucleus representing a different racial concretion from 
that first mentioned. One observes that in the upper Rhine Valley and 
in northwestern Switzerland, as in Lower Austria, this zone of extreme 
brachycephaly is broken, while a northern colony of it is found in Bavaria, 
Bohemia, and Silesia. 

In the Tyrol, southeastern Switzerland, and most of northern Italy is 
another nucleus, which is the home of the western branch of the Dinaric 
group, associated largely with the center of Rhaeto-Roman speech. These 
linguistic fossils are survivors of the pre-Germanic population of this 
region. Most of Austria itself runs longer headed, owing, no doubt, to 
the strong concentration of Germanic peoples there. The Dinaric region 
proper, extending from Bosnia to southern Albania, follows the mountain 
range, which in turn lies close to the Adriatic coast. The center of highest 
brachycephaly lies in southern Albania, in the Tosc country, well south 
of the center of tallest stature. The southern brachycephalic zone, of 
which it is the nucleus, extends far into Greece, along the western coast, 
from Epirus to the Gulf of Corinth. 

The curve of the Carpathians forms a brachycephalic barrier, within 
which all peoples represented, except for the Hungarian Szeklers, are 
very round headed. This infra-Carpathian brachycephaly pervades all 
other groups regardless of language, culture, or history. Beyond it lies 


the relatively long-headed expanse of the Polish, Ukrainian, and Molda- 
vian plain. 

As we turn to Asia Minor we see other instances of extreme regional 
brachycephaly. The Armenians, some of the Syrians, especially the 
Alouites, Lebanese, and Druses, are the roundest headed of all in this 
region. The Anatolian Turks, being typically brachycephalic, in this 
respect resemble modern representatives of the pre-Turkish peoples, of 
this region, notably the Armenians. 

The cephalic index map, like that of stature, shows that the Mediter- 
ranean Sea is by no means a racial unit. Some of the lowest and some of 
the highest cephalic indices in the world are found in close proximity to 
its shores. Another notable lack of continuity is seen in the far north. The 
hunting and fishing peoples, so consistently short of stature, are very 
variable in head form. The Lapps alone are consistently and extremely 
brachycephalic. The original mesocephalic head form typical of the 
Finns in their native habitat may still be observed in the regions occupied 
by Finnish survivals in central and northern Russia. 

On the whole, the distribution of the cephalic index in Europe and 
adjacent countries is extremely significant when one remembers the his- 
torical and archaeological background, but viewing its present distribu- 
tion alone one might easily form numerous false ideas about racial origins 
and continuities. It is sufficiently clear, however, that the zone of extreme 
brachycephaly in central Europe has several nuclei, and is separate from 
the Anatolian-Caucasic center and from that of the mongoloids of central 

One last factor remains to be mentioned, and this is the ultra-peripheral 
distribution of moderately high cephalic indices on the very westernmost 
fringe of Europe. One notices that southwestern Ireland has a mean 
cephalic index of 80 or over. Little spots of this same condition occur in 
northern Scotland, the Shetlands, the West Frisian island chain, in Feh- 
marn, and in points along the western Norwegian coast. This hypermar- 
ginal brachycephaly is peripheral to the dolichocephaly of northwestern 
Europe, which in its turn is a survival. The suggestion is that this round- 
headed tendency of the extreme western fringe is in the nature of a Palaeo- 
lithic reemergence. 

The third map of this series (Map 7), is intended to show the distribu- 
tion of absolute head size. Head size ideally should be a measure of the 
cubic capacity of the cranium, and capacity may be estimated upon the 
living by the use of the three dimensions, head length, head breadth, and 
auricular head height. Unfortunately, however, as already explained, 17 
auricular head height is for the most part an unreliable measurement, 

Page 243. 


331335 mm. 









and it would not be possible to construct a map covering a large area 
in which this was a component dimension. For this reason head size is 
here expressed simply by the sum of the length and breadth in each sample 
used. It so happens that large heads in the length-breadth sense are fre- 
quently high heads as well, so that there is little chance that the omission 
of the height dimension has falsified the appearance of head size condi- 

Head size is, in the first place, wholly unrelated to head form. Some 
of the largest heads are found among both dolichocephals and brachy- 
cephals, and the same is true of some of the smallest heads. It seems, 
however, to be closely correlated with total bodily bulk, and hence with 
weight, although not with stature. 18 This principle applies to other ani- 
mals as well as to man. Brain size is, after all, a component element of 
bodily bulk, and the requirements of the organism in the matter of nerve 
tissue depend apparently upon total size rather than upon the relative 
degree of attenuation of extremities. We have seen that cranial size is an 
important racial diagnostic in the cranium, and there is every evidence 
that it is equally important in the living. 

The map which shows the distribution of this trait is not, however, as 
reliable as the two which precede it. Lacunae have been filled in accord- 
ance with general racial trends and by the conversion of modern cranial 
material to living standards by fixed additions to allow for the soft parts. 19 
The areas which are least reliable are Portugal, Spain, much of France, 
and portions of western Germany. The Balearics and Sicily were filled 
in by inference. However, the data are sufficient to assure us that the 
general picture is correct, although the boundaries may well be inaccurate. 
The map will serve our purpose, and cannot lead us far astray, if we do 
not lean too heavily on it, or follow it in too much detail. 

The first impression that the map gives is one of a concentric distribu- 
tion of head size with Germany, Belgium, and northern France as the 
focus of greatest volume. From this focus bands of diminishing size stretch 
like bars dexter to the Persian Gulf. This pattern is broken in the Middle 
East by the intrusion of relatively large-headed mongoloid peoples from 
central Asia, and of non-mongoloid dolichocephalic Turkomans, Azer- 
baijanis, and Kurds. 

Studied in greater detail, where detail is justified, this basic pattern 
does not break down, but other facts appear. In the first place, Ireland 

18 Du Bois, E., CRIC, 1934, pp. 71-75; also, Marett, J., p. 129. 

19 Duckworth, W. L. H., JAPL, vol. 51, 1917, pp. 167-179. 
Fischer, E., MAGW, vol. 36, 1906, pp. 54-57. 
Gladstone, R. J., Biometrika, vol. 4, 1905/6, pp. 105-123. 
Mies, J., MAGW, vol. 20, 1890, pp. 37-49. 

Weisbach, A., MAGW, vol. 19, 1889, pp. 198-200. 


as a whole has the largest heads of any country excepting Belgium., A 
vertical line divides Ireland into a western, and especially southwestern 
half, with heads as large as the largest elsewhere, and an eastern, and 
especially northeastern, half with heads which although smaller, are still 
large by European standards. Iceland again is an area of maximum head 
size, and so are the Shetland Islands. Small regions of large head size 
appear along the Norwegian coast. The regions mentioned in this para- 
graph undoubtedly represent the maximum survival of Pleistocene Euro- 
pean man of the Briinn race in the northwestern portion of the continent. 
They coincide to a certain extent with the hypermarginal distribution 
of high mesocephaly and low brachycephaly. 

But there remains the bloc of large heads running from the Seine to 
East Prussia, and concentrated in Belgium and in the lower Elbe country. 
Here large heads are associated with brachycephaly, of varying degrees, 
but usually of a moderate order. This region has a much larger-headed 
population than has most of Sweden and Norway, and most of England 
and Lowland Scotland. The brachycephals of this large continental bloc 
all have head lengths which elsewhere go with dolichocephaly. The 
Fehmarn islanders, for example, whose small home is just south of the 
Danish Archipelago, have a mean head length of 193.5 mm., and a 
cephalic index of 83.6. 20 Their head breadth of 161.8 mm. is tre- 
mendous. In our historical chapters, we encountered but one racial 
type which consistently presented the combination of brachycephaly 
with great head lengths. That was the type found at Afalou and 
Ofnet, and in the Danish middens, and which was given the name 
Borreby. As will be seen later, the Borreby race has reemerged in the 
country where it was located during the Mesolithic and Neolithic 
periods, and it has become the most important single racial element in 
modern Germany. 

Palaeolithic and perhaps Corded survivals are to be seen in the large 
heads of the Finnish coast and northeastern Sweden; the track of German 
colonists in late mediaeval times is evident in Hungary and Rumania. 
The Basques have heads of considerable size also, and there seems to be 
a, significant nucleus of large heads in the Dordogne, where, as will be 
seen later, a long-headed, brunet Upper Palaeolithic survival seems in- 
dicated, as in west-central Wales. 

The zone of moderate head size lying between Germany and Poland 
on the one hand, and eastern Russia and the Caucasus on the other, 
seems to reflect an earlier Danubian and Nordic condition. In North 
Africa and southern Italy, small or medium-sized heads seem marginal 
and go with the older Neolithic Mediterranean element. The Hamites 

20 Sailer, K., Die Fehmarner, DRK, vol. 4, 1930. 


brought larger heads, such as are to be found today among Galla, 21 and 
among other predominantly Hamitic peoples. 

The tendency of the Hamites to large head size has divided the erst- 
while unified Mediterranean racial zone, which stretches across the whole 
lower quarter of the map, into a western and an eastern compartment. 
The eastern sector, from Cyrenaica to India, shows the small head size 
which apparently formed a cranial interlude in North African history 
between the end of the Capsian and the Hamitic invasions. As one leaves 
the map and passes into southern Arabia and Baluchistan, the heads grow 
smaller than any here designated. Here total length-breadth combina- 
tions of 328 mm. are found in the Hadhramaut and among Brahui. This 
zone which stretches along the northwestern shore of the Indian Ocean 
is part of the so-called Veddoid racial area, which does not extend into 
Europe or any region nearly approaching it. The racial character of the 
people inhabiting this zone can best be described in a more detailed 
chapter to follow. 

One of the most important results of the plotting of the head-size map 
is the discovery that the brachycephals of the white race and of Europe 
are not at all a unit in this respect, since they follow general racial zones 
which have no reference to head form. One may divide them into several 
sub-groups on the basis of head size alone. The Lapps, who in their pure 
form are hyper brachy cephalic, have very small heads. The other brachy- 
cephals of northern Europe, those concentrated in Germany, southern Den- 
mark, Belgium, and France, form the largest-headed group. These may be 
considered, tentatively at least, of Borreby derivation or inspiration. The 
Alpines of the Massif Central in France separate themselves clearly from 
this nucleus, with an emphasis on moderate head size. Although the re- 
gional data in France is poor, in this case it is sufficient to warrant the 
present conclusion. The Binaries are also moderate in head size, despite 
the coincidence of taller stature; only the Montenegrins themselves and the 
Albanians north of the Drin have truly large heads. The extreme hyper- 
brachycephals of southern Albania and Epirus are again of medium head 
size, like the Central French Alpines. The brachycephals of the Hungarian 
plain, and of the Carpathians, are for the most part also moderate. 

When we leave Europe and move to western Asia, we find that the 
Asiatic Binaries and the so-called Armenoids are in some areas smaller 
headed than the European Binaries; the Armenians themselves have 
heads approaching Binaric standards, but they vary regionally, with the 
largest heads in the northeast, toward the Caucasus. The brachycephalic 
Turks of Asia Minor are actually small headed, as are most of the Syrian 
brachycephals and the Iranian-speaking round heads of the Pamirs. The 

21 Unpublished data in author's possession. 


fringe of round heads along the southern Arabian, Persian, and Baluchistan 
coasts are very small headed, in a quite un-European sense. 

What are we to make of all this? The answer cannot be given as yet 
in final form, but several suggestions present themselves. 

(1) Head size, being a correlate of gross bulk, seems in general to be 
associated with regions of relative chill and humidity, all else being equal. 
The water content of the human body is greater where evaporation is 
least. In this way the flaccid Teutons and the fog-wreathed Irish in their 
moors and bogs have the heaviest bodies and the largest heads, while the 
indigo-stained Arabs, living on the utmost margin of desiccation, reach 
the opposite extreme in liquid economy. Man is not a water-storing crea- 
ture, like the cactus and the camel. 

(2) The largest-headed peoples are unreduced survivors or counterparts 
of Upper Palaeolithic man, who was a large-headed and presumably 
large-bodied animal. This applies both to dolichocephals and brachy- 
cephals. Brachycephaly is a mutative incident which may occur in any 
region or race, and head size may be more important than head form as 
an indication of ultimate genetic derivation, again all else being equal. 

It seems to me that somewhere between these two hypotheses lies the 
truth. Environment, which in the last analysis controls body size, must 
also eventually control the bulk of the head. But at the same time, genetic 
tendencies to absolute head size are inheritable, and without regard to 
head form. Hence early racial connections, under equal environmental 
conditions, may be better revealed by the size than by the shape of the 
vault. The heads of some people have remained constant in size and form; 
others have been reduced, brachycephalized, or both. But brachycepha- 
lization may take place without reference to body size, while reduction 
in head size is a corollary of general reduction. Here, as in general, the 
explanation of a given head size is an historical matter. 

Other criteria of the head and face would be difficult to plot. Face size, 
in general, is larger among the larger-headed and taller peoples of the 
northwest, and among those of rnongoloid affinity in the east. Most 
branches of the Mediterranean stock proper are characterized by rela- 
tively short and relatively narrow faces. The zone of long heads from 
Morocco to India is also a zone of small faces. This smallness, however, 
has as a rule no reference to the nose, which is one of the best racial criteria 
which we have, and one which is extremely significant. Unfortunately 
accurate charts cannot be made, since technical discrepancies render the 
use of statistics based on this organ almost useless in a large compilation. 

The nasal index among European peoples is typically leptorrhine or 
mesorrhine. The southern Mediterranean belt is typified by moderately 
leptorrhine peoples; and in the eastern extremity, where aquilinity is the 


rule, extreme leptorrhiny is very common. The most leptorrhine area in 
Europe itself is the Dinaric region, particularly Montenegro and northern 
Albania, where mean nasal indices below 60 are encountered. In most of 
western Europe the noses are leptorrhine, but when one moves into Russia 
and the northeastern Balkans, mesorrhiny becomes the predominant form, 
and nasal indices increase perceptibly as one moves eastward, to a high 
mesorrhine or even platyrrhine level. Turkish-speaking peoples in the 
East, however, form an exception to this rule. Turkomans, Azerbaijans, 
and the like are, as a rule, extremely leptorrhine, more so than the in- 
habitants of Asia Minor and the Caucasus. On the opposite side of the 
map, the extreme western fringe of tall, large-headed, meso- to brachy- 
cephalic peoples is likewise characterized by a slight increase in the nasal 
index. The Palaeolithic survivors were not notably leptorrhine ; they were, 
in fact, much less so than the Nordics and others who followed them. 

If one were to study the form of the orbits and the shape of the external 
eye, with adequate data, a very interesting and significant distribution 
might be seen. For example, the distance between the eyes is relatively 
great among all of the Slavic and Finnish peoples of eastern Europe, and 
this dimension increases as one approaches mongoloid territory. It is of 
moderate size in almost all of northwestern and central Europe, but 
again becomes pronounced in Ireland, along the coast of Norway, and in 
the Alpine regions, where one may attribute this wide-eyed condition not 
to mongoloid influences but again to a Palaeolithic survival. 

There are two zones of narrow inter-orbital diameters: (1) the entire 
Mediterranean zone from the Atlantic to India, and (2) the Dinaric zone 
reaching from the north of Italy to northern Greece. Again in the so- 
called Armenoid region of Anatolia and in Armenian territory itself, 
an extremely narrow inter-orbital distance prevails. This criterion may 
perhaps survive as a means of discrimination between facially character- 
istic Palaeolithic survivors and mongoloids, on the one hand, and basic 
Mediterraneans and Armenoid-Dinarics, on the other. 

The size, robusticity, and general form of the lower jaw is again an 
excellent racial criterion, but there is not enough data to permit it to be 
plotted. The Mediterranean zone from Morocco to India is characterized 
by a light, shallow jaw, a narrow bigonial diameter, and a restricted height 
dimension between the lower dental border and chin. This is the typical 
Mediterranean mandible, whether one finds it in Spain or in Arabia. The 
heaviest jaws and greatest bigonial diameters are found in the northwestern 
European borderlands, and in eastern Europe, where mongoloid influence 
is strong. The relatively light, narrow jaw of many Binaries and Armenoids 
again suggests that these types are for the most part brachycephalized 
forms of tall Mediterraneans. 



(c) Pigmentation, the Pilous System, and Morphology of the Soft Parts 

The fourth and last of the general distribution maps (Map 8), is de- 
signed to show the distribution of progressive degrees of blondism in the 
European area. While data on hair and eye color are plentiful, much 
material has been collected without the use of scales; although it is possible 
to correlate this with standard material in most major areas, the judgment 
of the compiler nevertheless plays a greater part here than in the maps 
which show the distributions of purely metrical characters. Under these 
circumstances, it has seemed most useful to divide the existing materials 
into five broad classes, designated and distributed as follows. 

The darkest stippling represents populations in which the hair is con- 
sistently black or dark brown (distinctions between these two shades are 
seldom valid), with less than ten per cent of a lighter hue. The accompany- 
ing eye color found in this brunet class is pure brown or black in over sixty 
per cent of the series; in most cases over eighty. Since all brunet white 
populations studied show some degree of mixed eyes 22 (green, blue, or gray 
in conjunction with brown), a small minority of this type seems endemic 
in the white racial stock, and must not be construed as evidence of racial 
blondism. Skin color, which again is an important element in blondism, 
varies less among Europeans than do either hair or eye color, and is more 
difficult to use. Hence it has been omitted from consideration in the 
draughting of the pigment map. 

The brunet hair and eye condition defined above, including a minimum 
of blondism, surrounds Europe and encroaches on all its borders, not ex- 
cluding the Atlantic. North Africa, almost all of Asia on or off the map, 
Portugal, most of Spain, southern Italy, Greece, the Aegean fringes, and 
finally, the northern pastures of the Samoyeds, converge to encircle the 
world's one important nucleus of blondism. 

The second most heavily stippled zone shown on the map, that of pre- 
vailingly brunet pigmentation, covers regions in which complete or partial 
blondism is not .rare, but is definitely less common than a purely brunet 
condition. The width of this zone depends, of course, upon the latitude 
of the category assumed by the author. In the present map, it is relatively 
narrow, and includes central and northern Spain, central Italy, most of 
the Balkans, the Caucasus, and a narrow vertical belt in eastern Russia. 
The Lapps, in their purest discoverable form, seem to fit into this class 
rather than the purely brunet one. Islands of prevailingly brunet pig- 
mentation occur far afield from the main zone, in parts of Wales, in 

22 The only valid exceptions seem to be the Ruwalla Bedawin and the Tuareg. See 
Chapter XI, sections 2 and 12. 









Morocco, Algeria, and Tunisia, in Crete, in the Jebel Druz, and in Luris- 
tan. The reasons for these exceptions are different in almost each case, 
and must be treated separately later. 

The decision as to the midpoint between blond and brunet hair and 
eye pigmentation hinges largely on one's definition of pure blondism. For 
practical purposes, pure eye blondism includes gray and blue eyes, with 
or without a small number of pigmented spots, or a narrow pigmented 
ring, near the pupillary border of the iris. It is impossible to segregate the 
spotted and unspotted in most data. Pure hair blondism includes, in the 
same arbitrary fashion, hair that ranges from light brown to ashen or 
golden. In the present map the intermediate class represents regional 
samples in which light and light mixed forms seem approximately equal 
to those which are prevailingly brunet. 

This intermediate zone is again narrow, and again continues the gen- 
eral scheme of concentricity. An exception to this scheme is seen among the 
Ostiaks, a Finnish group living along the banks of the Obi River and its 
western tributaries. Bulgars and Vlachs possess more blondism than a 
class four stipple would show, but hardly enough for the intermediate 
class when taken en masse. Therefore three capsules of intermediate 
stippling in the Balkans indicate these tendencies in a schematic manner. 
The northernmost and the southwesternmost represent concentrations of 
Vlachs, the middle one of Bulgars. 

Walloons of the province of Luxembourg, and southeastern Bavarian 
mountaineers, conversely, represent nuclei of intermediate pigmentation 
in blonder territory. One may postulate without difficulty that the 
Bavarian nucleus was once continuous with northern Italy through the 
Tyrol, for many Tyrolese are quite brunet, but the continuity has been 
broken by the Germanic advance in historic times up the Innthal. The 
refuge quality of the Austrian as well as of the Swiss Alps is conversely 
shown by the survival, since this Germanic thrust, of very blond local 
populations in the Lechthal and in other small, isolated valleys. As for the 
Walloons of Luxembourg they quite palpably represent a survival of pre- 
Iron Age brachycephals in their highlands, through the period of Celtic 
and Prankish invasions. 

The greatest difficulty of all in compiling this map lay in making the 
decision between what was predominantly blond, and what was merely 
more blond than brunet. If the eyes were almost uniformly light or light 
mixed, and the hair light brown or lighter in over fifty per cent of cases, the 
lightest group seemed indicated; if, in a majority of cases, the hair was dark 
or medium brown, or the eyes mixed, the second class was chosen. Some- 
times both hair and eyes indicated the second lightest stipple. In the pre- 
dominantly blond class, pure brunet pigmentation is less than ten per cent. 


The greatest degree of blondism recognized is definitely nuclear and, in 
fact, almost glacial in its distribution. There is, however, a nucleus within 
a nucleus; a center of lesser blondism which seems truly hyper marginal. 
This is the partial blondism of the Danish islands, of parts of the Nor- 
wegian coast, of Iceland, and of the southwestern tip of Ireland. This 
inner nucleus apparently coincides with the survival of the oldest, im- 
mediately post-glacial population. 

It is not unlikely that the original undifferentiated sapiens men, living 
in the Pleistocene, may have possessed a light brown or brunet white skin 
color, with black or dark brown hair and brown eyes. Different racial 
stocks which grew out of this common base by differentiation, mixture, or 
both, may have shown early tendencies to develop specialized variations 
of their own in pigmentation. Such tendencies are likewise seen within 
single species of ape, such as the gibbon, chimpanzee, and gorilla. The 
negroid races, for example, must have formed, before the end of the 
Pleistocene, a progressive tendency toward an abundance of dense pig- 
ment cells in the skin, the fundus, and the iris; while the whites, before 
their dispersal from a common center, must already have developed a 
tendency, presumably recessive, toward blondism. The universality of 
some degree of blondism among whites and near whites everywhere 
makes it unlikely that it was ever confined to a single race or group of 
races within the white family. 

Blondism is a state of partial depigmentation, due to the paucity of 
melanin granules in the skin, hair, and iris, and, with some types of pig- 
ment, to the small size of these granules. The pigment granules are com- 
posed of a substance known as melanin, the chemical composition of which 
has been roughly determined. 23 Melanogenesis, the process by which 
melanin is formed, is "an intercellular enzymic oxidation process, in which 
an amino acid chromogen is converted, with the aid of catalytic copper, 
to the pigment melanin." 24 It has been, proved by experiments with rats 
and rabbits that a dietary deficiency in copper produces a pigment reduc- 
tion, 26 and that with the restoration of a normal diet, the animal's normal 
pigmentation will return. Hence blondism, being a phenomenon of pig- 
ment reduction, is presumably caused by a genetically controlled limita- 
tion of the oxidation process dependent upon the body's supply of copper. 

23 Melanin is approximately 55 per cent carbon, 6 per cent hydrogen, 12 per cent 
nitrogen, 2 per cent sulphur, and 25 per cent oxygen. Young, W. J., BJ, vol. 15, 1921, 
pp. 118 seq. 

24 Glodt, H. R., Melano gene sis, a thesis submitted for honors in Anthropology at 
Harvard University, April, 1937. MS. in Peabody Museum Library, Harvard Uni- 
versity. Quotation from p, 71 . Author's permission secured. This whole section is taken 
largely from Glodt. 

25 Cunningham, I. J., BJ, vol. 25, 1931, pp. 1267 seq. 


Blondism therefore may have originally been motivated as a response to a 
mineral deficiency through an endocrine agency of control. There is no 
reason now known why it should be limited to whites, but actually its 
appearance among members of other major racial groups is rare. 

Although skin color is apparently a directly quantitative matter, hair 
color, it is now known, is determined by two different pigment factors. 
One is composed of oval or spindle-shaped cells of melanin, of varying size 
and frequency. 26 When these cells are large and overlap, within the trans- 
lucent body shaft which lies between the central canal and the outer 
horny layer of the individual hair, the hair appears black or dark brown. 
When the cells are smaller they appear yellowish or light brownish, al- 
though the chemical composition of the melanin is the same. The size of 
these cells, therefore, and their abundance within the hair cortex, deter- 
mine the degree of blondism or brunet coloring. 

The second pigment factor which influences hair color is rufosity. Red 
hair contains a fine stain, at first considered amorphous, which is now 
thought to be composed of extremely fine cells, probably slightly different 
in molecular structure from ordinary melanin. 27 This stain may be present 
or absent, and if present may be faint or intensive. Thus it is both qualita- 
tive and quantitative in reference to ultimate hair color. If it coincides 
with large, dark melanin cells, the black color so caused may mask the 
rufosity in all but unusual lights, while if a large amount of it coincides 
with blondism, red hair is the result. It is likely that golden hair is caused 
by a combination of blondism with a slight degree of rufosity. 

If one could test for rufosity accurately with all pigment shades, it 
would probably be seen that this character has no association whatever 
with blondism, but is a purely independent variable. That this is likely 
is shown by the fact that rufosity is completely uncorrelated with eye 
color. 28 Thus rufosity may be wholly absent in many normal individuals, 
while the melanin cells are totally absent only in albinos. Rufosity may, 
by the same token, be lacking in entire races, and with better data it 
might be possible to discover the racial significance, if any, of this ap- 
parently functionless condition. Within the blonder segment of the white 

26 Coniteer, H., ZFMA, vol. 29, 1931, pp. 83-147. 
Hausman, L. A., AJPA, vol. 12, 1928, pp. 276-277. 

Jankowsky, W., ZFRP, vol. 5, 1932, pp. 1-48, 111-119; also VGPA, vol. 6, 1931-32, 
pp. 66 T 69. 

27 C^Miitzer, H., op. cit. 

Klinke, K., BZB, vol. 160, 1925, pp. 28 seq. 

28 Conitzer, H., op. cit. 

I have separately confirmed this claim by making 1 30 contingency tables, of six or 
more boxes, between hair and eye color, in each of which a negotiable amount of ru- 
fosity was present. In every instance red hair was found to be completely complacent 
to eye color. 


race, however, we know that rufosity has a regional and a racial connota- 
tion. Blond hair is readily divisible into two categories, golden and ash- 
blond (cendr6), which are distinguished on the Fischer hair color chart. 
Light brown and brown hair shades similarly may be segregated on the 
same basis into two separate and parallel classes. 

The pigmentation of the iris is more suited for refined analytical study 
than either skin or hair color. Skin tans and weathers, while hair bleaches 
with the sun and darkens with advancing age, until the advent of graying; 
the iris, on the other hand, retains its pigment pattern with relatively little 
change. If studied under constant light conditions, so that the pupil is 
contracted and the concentric muscle zones flattened, the iris is seen to be 
a detailed field of muscle-layers and pigment cells, of considerable com- 

In all but albino tic eyes, the inner wall of the iris is permeated with 
melanotic pigment cells so overlapped as to make the iris, whether dilated 
or contracted, a perfect light-proof diaphragm. It is this pigment lining, 
reflected through several layers of outer iridical tissue, that gives a light 
eye its blue appearance. Additional pigment presents its true brown color. 
Thus in a mixed eye of complex pattern it is possible to plot the relative 
depths of different groups of pigment cells. Cells concentrated along the 
radial, dilating muscle fibers give the eye a rayed appearance, while 
those lumped about the concentric sphincters produce a zoning. In a 
black eye the surface pigment is so dense that it is impossible to see into 
the iris, but in a brown eye it is usually possible to make out some of the 
pattern. Many purely brunet eyes show a contrast between different 
brown-producing layers. 

In purely light eyes, in which no surface pigment is seen, there are 
nevertheless differences in coloring v hich are readily noticeable and which 
may be used as criteria of racial differentiation. The principal distinction 
is that between the blue eye, which in its extreme form takes a deep sky- 
blue color, and the gray eye, which in its extreme form is almost white. 
Since these two forms grde into each other without a natural line of de- 
marcation, the factor which distinguishes them must be considered quan- 
titative rather than qualitative. Research on this subject does not seem as 
yet to have been done; we do not know what causes this difference, and 
can only repeat Bryn's speculation that it has something to do with the 
relative coarseness and opacity of the radial iris muscles, through which the 
pigment in the posterior walls of the iris is reflected. 29 

Geographically and in individuals, it is possible to make valid correla- 
tions between the four end types of hair and of eye blondism. The golden 
type of hair, whether blond or brown, tends to be associated with* the 
^Bryn, H., Homo Caesius, p. 19. 


bluer shades of eye, whether pure light or mixed; on the other hand, the 
ash blond type of hair usually goes with a grayish iris. At present there 
seems to be no direct reason for these linkages, but we have much to learn 
about these matters. 

At any rate, when we apply this distinction to the map, we see that the 
golden-blue combination is commonest in the western half of our nuclear 
zone of light pigmentation, in Norway and the British Isles; while the 
ashen-gray combination is more typical of Sweden and of the lands east 
of the Baltic. In the western half of the blond nucleus, and especially in its 
British periphery, there is an asymmetry of linkage, for in Ireland, for 
example, a world's extreme ratio of light eyes is associated with hair which 
is often brown or dark brown. On the eastern side the opposite is true; in 
Poland and southern Russia ashen hair of a very light shade goes fre- 
quently with dark-mixed or brown eyes. These regional asymmetries 
weaken the total unity of blondism, but do not destroy it. 

From further correlations between types of pigmentation and other char- 
acters, such as stature, bodily build, head size, head form, and face form, it 
is possible to show that the golden-blue variety, with rufosity, is partly as- 
sociated with the old Palaeolithic hunting strain, while the ashen-gray ex- 
treme goes rather with the Iron Age Nordic range of types, and with eastern 
European blonds of various degrees of superficial mongolism. Within his- 
toric times the zone of frequent blondism stretched from north western Eu- 
rope across the Russian steppes into central Asia where it touched China, 
but violent and rapid ethnic movements in Asia have nearly eliminated this 
eastern extension. We do not know how long ago the distribution map of 
blondism assumed its present concentric and glaciation-like character. 

It is very probable that pigmentation is definitely capable of alteration in 
response to environment, through selection. Blonds in the tropics are at a 
disadvantage, particularly if living under primitive cultural conditions. A 
black skin with a profusion of sweat glands, like that of the African negro, 
must be better than a pinkish integument which is subjected to repeated 
burning and blistering, and incapable of tanning. 30 In the iris, the pigment 
in the posterior wall acts as a completely light-proof diaphragm, and hence 
there can be no direct functional disadvantage to a gray or blue iris, as with 
thatof an albino. But since the iris color seems to be, as Wilmer has shown, 81 
correlated with the pigmentation of the retina, eye blondism may serve to 
indicate the presence of a functional disadvantage. It is conceivable, but 
not as yet demonstrable, that the chocolate-brown pigment cells in the 
negro's fundus may give his optic nerve more cc ;afort in the desert glare 
than the pinkish, almost pigmentless retina of tru, blond white man. 

80 Baur, E., Fischer, E., and Lenz, E., Human Heredity, p. f A. 
* Wilmer, W. H., Atlas Fundus Oculi. 


Black skin and a black eye, then, may be variables which are advan- 
tageous under hot, bright, equatorial light conditions. A partially depig- 
mented skin and fundus condition can perhaps survive without disadvan- 
tage only in a climate where the light is weak. Blond hair, however, cannot 
be assigned any survival value of either a negative or a positive character. 
Until definite experimental evidence is at hand, we must postulate that 
only through its partial genetic linkage with skin and eye color is the 
blondism or darkness of hair determined. On the whole, the totality of 
evidence in regard to blondism as a unit indicates that this phenomenon 
is a recessive trait endemic in the white racial stock, and that it has be- 
come a major racial character only among groups of people living at one 
time under light conditions of sub-glacial intensity. This applies to the 
Upper Palaeolithic strain in part or as a whole, and to certain of the more 
northerly Mediterranean branches. The mongoloids and American In- 
dians living under parallel conditions apparently lack the initial mutative 
tendency necessary for its development. 

In the European zone of maximum blondism are included tall and short 
populations, long-headed and round-headed, eagle-beaked and snub- 
nosed; many such variations occur to which degree of pigmentation seems 
complacent, Within the two main types of blondism, racial sortings are 
clearer, but on the whole blondism alone assumes the character of an 
unlinked mutant. 

Without actual maps, there is little use in reviewing the distribution of 
the pilous system and soft parts, in more than a cursory manner, since 
these will be discussed at greater length in the chapters to follow. Hair 
form, which according to Haddon is the most important racial criterion 
to be found in man, is of little use in distinguishing white sub-groups. 
Most European hair is straight or slightly wavy, although exceptional in- 
dividuals in the straightest-haired groups have ringlet forms. Curly hair 
of this description is quite common in western Ireland and in Wales; it is 
also frequent in the whole of North Africa and in the western Mediter- 
ranean shorelands of Europe. Eastern Europe is predominantly straight 
haired, and as one approaches mongoloid territory this condition of course 
becomes more pronounced. 

The amount of body hair on the adult male is closely correlated with the 
amount of beard, and both are linked with age, for a hairy man grows 
hairier as he becomes older. At the same time, baldness is most frequent 
among those with heaviest body hair and heaviest beards. Browridges, 
and other bony excrescences of a hypermasculine nature, are closely linked 
with excessive pilous development of the body and beard, and with a 
tendency to baldness. Europeans, on the whole, are among the hairiest- 
bodied and heaviest-bearded groups of men, being equalled or exceeded 


only by the Australians and the Ainu. Both negroid and mongoloid skin 
conditions are inimical to excessive hair development except upon the 

The Mediterranean peoples, on the whole, are less hairy than other 
Europeans. Pure dolichocephalic Europeans, of normal Mediterranean 
type, whether blond or brunet, tend to a hairless chest and a patchy beard. 
Among Arabs a complete beard is rare, and is considered a sign of evil 
character. One must look upon great hairiness, and a great beard de- 
velopment, as well as a high incidence of baldness, as a multiple endocrine 
manifestation associated with relatively great sex differentiation in a 
masculine direction. Alpines and Central Europeans, in general, show an 
excess of this combination, and so do many Balkan peoples and Near 
Eastern Asiatics. This combination is in Europe associated with the non- 
Mediterranean element in the composition of the white stock, although in 
Asia the cleavage is not so clear. The baldness which is part of this com- 
plex is of genetic motivation, and differs in cause from the dry-scalped, 
fine-haired alopecia associated with extreme hair blondism. 

The morphology of the external eye is also subject to regional distribu- 
tion. High orbits, with no folds, are characteristic of Binaries, and of most 
Near Eastern peoples; orbits of moderate height, and with a tendency to 
external folding in maturity and old age, go with long-headed peoples of 
both blond and brunet varieties, while a median fold, indicative of both a 
low orbit and a thick fat deposit in the eye region, goes rather with the 
Finnic and Slavic blond mesocephals and brachycephals. The true inter- 
nal or mongoloid fold is not common in Europe and is found in numbers 
only in the east, in the Kalmuck and Tatar districts of Russia, and in the 
far north. 

Extreme cragginess and ruggedness of facial features, including the 
forehead, the superciliary region, the malars, the jaws, and the nose, are 
associated with the western marginal fringe area, and especially with the 
region of largest heads and maximum Palaeolithic survival. Nordics and 
Mediterraneans, whether in Europe, North Africa, or southwestern Asia, 
have a maximum of facial relief, without this appearance of bony massive- 
ness. The malars are laterally compressed, the nose thin and often beaked. 
Facial flatness, intensified by fatty deposits over the malars, while more 
typical of mongoloids, becomes characteristic in eastern Europe and ex- 
tends into Poland, Finland, and Hungary. 

The maximum nasality of the Near Eastern peoples, of whatever head 
form, is accompanied by a number of related features. One of these is the 
concurrency of the eyebrows over the nose, which is geographically cen- 
tered in the Near East. Another is the predominant convexity of the nose 
as a whole, and the depression of the tip, especially in old age. In man the 


nose passes through a definite and continuous cycle of growth changes 
comparable in form, if not in degree nor in exact anatomical detail, to 
those found in the proboscis monkey. The nearest approach to the pro- 
boscis in the extent of nasal change is, however, found among Near 
Easterners from Armenia to Afghanistan. In Europe the same is true to 
a lesser degree in Albania and Montenegro. 

A map showing the form of the nasal profile would have centers of con- 
vexity in the Dinaric area and throughout western Asia, with the excep- 
tion of Arabia; centers of concavity would lie in the north of Scandinavia, 
and across the whole of eastern Europe from the Baltic onward. The rest 
of the map would be relatively undifferentiated, with all forms present, 
but the straight profile most common. 


We have reviewed some of the characters upon which race, in the 
sense of sub-divisions of the so-called white branch of living humanity, 
may be classified. It has become apparent from this review, as from the 
earlier chapters, that the "white" racial family is a composite amalgam- 
ation of peoples thrown together by accident of geography, blended into 
some semblance of homogeneity in major diagnostic features, and altered 
by environmental and cultural circumstances and by migration. Before 
attempting to propose a classification of living whites, however, it may 
be wise to pass in brief review the more important or more influential 
theories by which race has been classified in the past. The history of 
racial classification is a subject for a book in itself, and here we propose 
to limit our discussion to its minimum. 

It is impossible to say when man began to classify himself into races. 
Knowledge of racial differences must, however, be as old as these differ- 
ences, and must from the beginning have been a factor in their develop- 
ment. The Egyptians were well aware, of the racial problem, and took 
pains, in their art, to differentiate between the various kinds of men that 
they knew. The Greeks likewise made classifications; both Hippocrates 
and Aristotle were strong environmentalists, as were the mediaeval Arab 
geographers who followed the classical tradition. Since these ancients 
and mediaevals wrote before the discovery of the bell-shaped curve (normal 
probability curve), probable error, correlation, or of the cephalic index, 
their system of classification was both observational and intuitive, and 
operated by the mechanism of generalization. Despite the work of the 
biometricians, and the mechanization of physical anthropology during 
the last half century, all important or influential systems of classification 
yet devised still operate on the same principle. . 

Aside from the ancients and their mediaeval followers, the modern 


period in physical anthropology begins with Blumenbach, whose system 
is still employed by most grade school geographers. Blumenbach, some 
one hundred and fifty years ago, divided mankind into the familiar white, 
black, brown, yellow, and red races, basing his primary classification upon 
skin color, although he considered other characters as well. In breadth 
of popular acceptance, his is still, in its simplest form, the most influential 
classification. During the first half of the nineteenth century, the world 
of science as well as the public was inclined to accept Blumenbach' s divi- 
sions without too much protest, but in the period from 1860 to 1890, 
Europe was rife with attempts to classify mankind into orderly sys- 
tems. 32 

In 1878 Topinard proposed a classification based not on skin color, but 
on hair form. 33 Haeckel 34 and Miiller 35 proposed the same diagnostic one 
year later. Topinard did not, however, rely on one character alone, but 
included skin color and nose form as subsidiary diagnostics. During this 
general period of activity systems were proposed by such varied authorities 
as Huxley, Geoffrey de St. Hilaire, and de Quatrefages; 36 it was the last 
named, who, during the war of 1870, first prostituted the materials of 
physical anthropology for the purposes of nationalistic propaganda. The 
gauntlet flung down by de Quatrefages, who called the Germans "Huns," 
was seized by his enemy and converted into the more effective weapon of 

It remained for another Frenchman, however, to coin the word "Nor- 
dic." This was Deniker, 37 who has had a greater influence upon the sub- 
sequent classification of race than any of his nineteenth century contem- 
poraries, and who still remains the most important classifier. In view of the 
lack of scientific method available at his time his classification, later 
modified slightly, was first published in 1889 his intuitive genius and 
his grasp of patent situations were extraordinary. Born in Russia and 
educated in St. Petersburg as an engineer, he had travelled widely through- 
out eastern Europe and the Caucasus before he settled in Paris in 1876, 
at the age of twenty-four, to begin his career as an anthropologist. 

The first step in Deniker's system was to divide mankind on the basis 
of a combination of hair form, hair and eye color, and nose form, with 
hair form as the principal diagnostic. He made six primary divisions, as 
shown on page 281. 

82 Deniker, J., The Races of Man, pp. 280-284. 
88 Topinard, P., RDAP, second series, vol. 1, p. 509, 1878, etc. 
84 Haeckel, E. H., Naturliche Schopjungsgeschichte, vol. 7, pp. 626, 647. 
^Muiler, Fr., Allgemeine Ethnographie, pp. 17-19. 
38 Deniker, J., loc. cit. 

37 Deniker, J., BSAP, vol. 12, 1899, p. 320; JRAI, vol. 34, 1904, pp. 181-206; Th$ 
Races of Man, pp. 285-286. 


A. Woolly Hair, Broad Nose. 

B. Curly or Wavy Hair. 

C. Wavy Brown or Black Hair, Dark Eyes. 

D. Fair, Wavy or Straight Hair, Light Eyes. 

E. Straight or Wavy Hair, Dark, Black Eyes. 

F. Straight Hair. 

Within these primary divisions he based his further classification upon 
combinations of skin color, nose form, stature, cephalic index, pilous de- 
velopment, browridges, and other characters. By this means he properly 
separated the Bushman into a separate class of group A; the Australian 
as well as the Dravidian went into group B, and were thus separated from 
a major "black race." Within the straight-haired class the Lapps, Mongo- 
loids, and American Indians were arranged into what seemed then a 
reasonable order. 

The white group, with which alone we are concerned, falls almost en- 
tirely into his C and D categories, with one segment in B. The table on 
page 282 gives this section of his classification in full. 38 

In that table, Deniker lists a number of races found both in Europe 
and outside that continent, of which eleven, not counting the Ainu, 
might be classified as basically white. His #5, the Ethiopian, is the 
Hamitic race of East Africa, with or without a slight negroid increment; 
his #9, the Indo-Afghan, is the hook-nosed type of Mediterranean which 
we have found to extend from Mesopotamia to India across the highland 
belt, from at least the third millennium B.C. onward; his #8 is Armenoid. 
The inclusion of the Ethiopian and Assyrioid with Australians and Dravid- 
ians rather than with whites, while inexact, points, in the first case, to 
the negroid admixture of modern Ethiopians, and in the second, to a 
realization of the affinities of Australoids and Veddoids to the white group 
as a whole. 

Besides these three, in effect, Hamitic, Armenoid, and Irano-Afghan, 
he finds two other white races outside the continent of Europe proper: 
these are his Arab and Berber. Thus we find a total of five morphologi- 
cally white races in Asia and Africa; four of these are actually sub-divisions 
of the cranially unaltered basic Mediterranean stock. 

In Europe itself he finds six races; the Littoral European, also called 
Atlan to-Mediterranean, is the tall Mediterranean associated in antiquity 
with Megalithic cultures, and may be related basically to Deniker's 
Ethiopian. His Ibero-Insular is the short Mediterranean race of Spain 
and the western islands, and corresponds to the Neolithic Mediterranean 
type in these regions. Deniker distinguished, therefore, between certain 

88 Based upon Deniker' s 1912 classification with some reference to his 1889 scheme 
as well. 




Reddish-brown, narrow nose, j 

tall stature, dolichocephalic. J 

Chocolate-brown, broad nose, 1 

medium stature, dolichocephalic. J 

Brownish-black, broad or narrow j 

nose, short stature, dolicho- I 

cephalic. J 

2. SKIN OF tawny white, nose narrow, 

hooked, with thick top, brachycephalic. 

1. DARK 


1. SKIN clear brown, black hair, narrow, \ 

straight or convex nose, tall \ 

stature, dolichocephalic. J 



Dravidian (sub-races 
Platyrrhine and Lep- 


Indo- Afghan 

Aquiline nose, promi- 
nent occiput, dolicho- 
cephalic, elliptical 
form of face. 

Arab or 


2. SKIN 


Straight, coarse nose, j 
dolichocephalic, i 
square face. J 


(4 sub-races) 



Straight, fine nose, 
mesocephalic, oval face. 

( A tlan to- 



Short stature, dolichocephalic 




Short stature, strongly brachy- 1 
cephalic, round face. j 




Tall stature, brachycephalic, 
elongated face. 

) Adriatic 
j (Dinaric) 



SKIN REDDISH ( Somewhat wavy, reddish, tall 
stature, dolichocephalic. 

> Nordic 


Somewhat straight, flaxen haired, 
HAIR FLAXEN s ^ ort stature, sub-brachycephalic. 

(R. Orientals) 



SKIN LIGHT BROWN, very hairy body, broad and concave A . 

i i i i i Ainu 

nose, dolichocephalic. 



of the basic sub-varieties of the Mediterranean family, and except for 
the categories Arab and Berber, this distinction is on the whole accurate. 
He was aware of the differences between the three most important sur- 
viving divisions; (a) Short Mediterranean, (b) Tall, Megalithic, and East 
African variety, and (c) Hook-nosed, Indo- Afghan or Irano- Afghan variety. 

At the same time, he was aware of the distinction between the Alpines 
and Binaries, both in form and in geographical distribution. In his 
placing of the blonds into a separate category, he was following a taxo- 
nomic system rather than an estimate of relationships. His Nordics are 
accurately defined on the basis of living peoples; they are given a cephalic 
index of 77 to 79, instead of a non-existent lower mean; and they are 
segregated from the blond brachycephals of central and eastern Europe. 

In order to accommodate other racial elements not fully covered by 
these classes, Deniker devised certain sub-races: (1) The Northwestern 
sub-race, a division of the Atlanto-Mediterranean, to accommodate es- 
pecially the dark-haired western Irish. (2) A Sub-Nordic, which differs 
from the Nordic in the possession of mesocephaly, a square face, and a 
turned-up nose; this was devised to accommodate peoples living to the 
east of the Baltic and in northern Germany. (3) The Vistulan race is a 
branch of the eastern European or Oriental. The Oriental is described as 
short statured (163-164 cm.); moderately brachycephalic (GJ. = 82-83); 
and possessing light yellow or flaxen hair, a square cut face, a nose which 
is frequently turned up, and blue or gray eyes. This race is associated 
with the eastern Slavs and Finns for the most part, while the Vistulan is 
a variety of the same race with shorter stature and mesocephaly. The 
last of Deniker's secondary races is the Sub- Adriatic, described as a slightly 
shorter, slightly less brachycephalic and blonder variety of Dinaric, with 
a stature of 166 cm., a C. I. of 82-85; and derived from a blend of Dinaric 
with Sub-Nordic. 

Two other authorities of what might, be called the prestatistical school 
deserve mention at this point Sergi and Ripley. Sergi, 39 whose main 
interest was the Mediterranean race, based his classification primarily 
upon the circumferential profile of the head when seen from above, and 
worked more with crania than with the living. His chief contribution 
was the realization of the basic unity of the Mediterranean race, in both 
its blond and brunet forms, and its connection with the bearers of Euro- 
pean civilization. Thus he anticipated the findings of the archaeologists 
that the Neolithic economy was brought into the western world by Med- 

He also made it clear that the so-called Brown Race, in its dolicho- 
cephalic and leptorrhine or mesorrhine forms, was for the most part an 

89 Sergi, G., Specie e vaneta umane; UUomo; Le Origini Umane; The Mediterranean Race. 


extension of the same Mediterranean family into southern Asia. He divided 
whites into Eurafricans, which is another word for basic Mediterraneans, 
and Eurasiatics, under which he included all brachycephals of white affin- 
ity. Sergi anticipated the discovery not only of the unity and cultural im- 
portance of the Mediterraneans, but also the dual origin of the white race. 

If the schoolchildren and the unerudite public at large still follow 
Blumenbach, and the anthropologists themselves devise classificatory 
schemes based upon Deniker, the large intermediate group of educated 
laymen rely almost entirely upon Ripley. 40 Ripley, writing in 1899, was 
aware of Deniker 's work, but rejected it. He considered that Deniker 
had made the picture much too complicated, and that there were but 
three white races, the Teutonic (Nordic), the Alpine, and the Mediter- 
ranean. The Nordic and Mediterranean were old European branches of 
an earlier white stock, while the Alpines were immigrants from Asia who 
had brought agriculture and the whole Neolithic economy with them. 
The Alpines, besides introducing a new physical type, parted the Nordics 
from the Mediterraneans geographically, so that the two might develop 
separately, and that the Nordics in particular might derive their tall 
stature and blondism from environmental causes in isolation. 

The above brief exposition has many advantages. It is simple, it is 
lucid, it is easily remembered. It fitted into the linguistic picture of Aryan 
culture bearers plodding across Europe from their simple home in the 
Hindu-Kush, developed by nineteenth century philologists, although 
Ripley himself was vehement in his rejection of linguistics as a proper 
approach to racial study. At the same time it explained the newly-found 
and well-preserved Neolithic remains of the Swiss lake dwellings. 

With such a simple scheme, it was easy for Ripley's followers to tack 
psychological characters to the three-fold framework, and the "Nordic 
with a genius for leadership and government," "the stolid, unimaginative, 
plodding but virtuous Alpine," and the "gay, artistic, and sexy Mediter- 
ranean" soon followed. Hilaire Belloc's famous verses, published orig- 
inally in the New Statesman^ satirize this attitude perfectly. 

"Behold, my child, the Nordic man, 
And^ be as like him as you can : 
His legs are long his mind is slow 
His hair is lank and made of tow. 

"And here we have the Alpine race. 
Oh ! what a broad and brutal face. 
His skin is of a dirty yellow 
He is a most unpleasant fellow. 

* Ripley, W. Z., The Races of Europe. 


"The most degraded of them all 
Mediterranean we call. 
His hair is crisp and even curls 
And he is saucy with the girls." 

Ripley himself had little or nothing, in a direct sense, to do with this 
efflorescence of speculative psychology, for the attitude of differential 
racial values had been crystallized as early as Gobineau; 41 but he did 
give the exponents of this school a facile terminology. Racial nationalism 
had been growing before Ripley' s time; but he, for the first time, gave the 
laymen a racial classification which they could understand, and which 
could be converted into catchwords. 

Like his predecessors, Ripley was discreet about the age of white men 
on European soil; only in the case of the Alpines was he willing to set a 
culturally stabilized date. In his day it was generally believed that the 
Neolithic went back to anywhere from eight to fourteen thousand B.C., 
and the Mesolithic period was not generally recognized. Furthermore 
the function of the glacier in regard to human habitat was but poorly 
comprehended. Ripley did, however, make one speculation about the 
survival of preglacial man in Europe; he postulated that some of the 
inhabitants of the Dordogne region in France might be Cr6-Magnon 

Some twenty years previously Verneau ^ had remarked upon the re- 
semblance between Guanche crania from the Canary Islands and these 
Cr6-Magnon skulls, and had postulated a genetic relationship between the 
two peoples so separated in space and in time. In 1896 von Luschan and 
Meyer 43 reaffirmed this relationship, and this endorsement prepared the 
way for a more accurate realization of the part played by survivors from 
the last glacial period in the modern peopling of Europe. It was soon 
realized that, if Upper Palaeolithic man could survive in the Canary 
Islands, he could persist elsewhere as,well, and from this start arose the 
theory that the Crd-Magnon people had retreated northward with the 
glacier, and had survived in Scandinavia. Paudler, 44 in his Die kellfar- 
bigen Rassen, first put this thesis into digestible form, and distinguished 
between his "Dalo-Nordic" or "Palish" (Gunther), which is tall, long- 
headed, with a mesorrhine nose and short, broad face, and a "Teuto- 
Nordic" which is also tall and long headed, but has a long, narrow nose 

41 Gobineau, A. de, Essai sur Vinegalite des races humaines. 

42 Verneau, R., BMSA, Paris, ser. 2, vol. 2, 1876, pp. 408-417; Arch des Missions 
Scientifiques et Litteraires, Paris, 1887, ser. 3, vol. 13, pp. 567-817. 

48 Meyer, H., Die Insel Tenerijfe; Uber die Urbewohner der Canarischen Inseln. 
Luschan, F. von, article in Meyer, Tenerijfe. 

44 Paudler, F., Die hellfarbigen Rassen. See also his earlier article in Anthropos, vols. 12- 
13, 1917-18, pp. 641-694. 


and face form. The first is considered to be the primary Cro-Magnon 

From this thesis has arisen the idea, in conjunction with philology and 
archaeology, that the Germanic peoples, as descendants of Cr6-Magnon, 
represent the racial and linguistic nucleus of the Indo-Europeans; that 
European Neolithic civilization and Indo-European speech both had 
their origin in northern Germany and Scandinavia; that the Corded 
people, a Nordic variety, originated and spread from here; and that in 
effect, the Nordic race, Indo-European speech, and European culture in 
its basic form, arose from Palaeolithic racial and cultural origins in this 
northwestern European glacial center. This theory, bolstered on the 
archaeological side by Kossinna, 45 is popular in Germany, but is by no 
means endorsed by all German physical anthropologists. 

The modern German school has made a great advance over Deniker 
and his contemporaries, and over Ripley, in the realization that an im- 
portant element in the modern European racial conglomerate is of glacial 
antiquity in Europe. The difference between their conclusions and those 
of the present study lies mainly in my acceptance of Childe's derivation 
of the Neolithic economy, and Menghin's as well, rather than that of 
Kossinna. Von Eickstedt, 46 the most articulate of the modern German 
raciologists, in his derivation of European peoples from Asia at various 
periods, does not emphasize the introduction of the food-producing 
economy in this connection. 

It would be outside the scope of the present study to attempt a complete 
survey of current ideas and current classifications which concern the 
European races. A partial survey would, on the other hand, be unfair to 
those who might, through limitations of space, be neglected. 1 shall, 
therefore, limit my exposition to the systems of two authors, 47 von Eick- 
stedt and Czekanowski, who have been particularly occupied with the 
question of racial taxonomy and who are the most vocal members of the 
German and Polish bodies respectively. Their influence has been con- 
siderable, and their schemes are articulate and orderly. 

Von Eickstedt, whose Rassenkunde und Rassengeschichte der Menschheit 
represents the most ambitious attempt at world-classification yet made, 
follows, in his European sections, three masters: Ripley, Deniker, and 
Montandon. It is the combination of these three, skilfully blended, 
which has produced his system. In the first place, he agrees with Ripley 
that there are but three basic races in Europe; Nordic, Alpine, and Med- 
iterranean. These three are typically confined to three climatic and 

46 Kossinna, G., Ur sprung und Verbreitung der Germanen, MannusB, #6a, 1928. 
46 Eickstedt, E. von, Rassenkunde und Rassengeschichte der Menschheit. 
47 1 am omitting Giinther, despite his great vogue, since his system is a close Germani- 
zation of Deniker's, with a few changes. 


geographical zones; the cold northern plain, the central mountain belt, 
and the warm belt reaching along the Mediterranean shores, and over 
Arabia and Iran to India. 

He differs from Ripley, however, in that he divides his three zones into 
sub-races, and here he follows, for the most part, Deniker. The northern 
zone is occupied, at its western extremity, by the Nordics; at its eastern 
by his Osteuropid race, the Orientale of Deniker, and the East Baltic of 
Nordenstreng and of authors writing in English. 48 The central mountain 
belt is occupied, reading from west to east, by the Alpines, the Binaries, 
and, in Asia, the Armenoids, and the Turanids, the latter being the leptor- 
rhine brachy cephalic central Asiatic Turkish racial form. The southern 
zone is occupied by the Mediterraneans on the west, then the Orientalids 
(Beniker's Indo-Afghan) in North Africa, and thence over to Khyber 
Pass, where the Indid race begins. 

In the differentiation between the segments of each zone, Montandon's 
ideas, 49 elaborated from those of Rosa, come into play. Von Eickstedt, 
following the principles of the ologenesis theory, has decided that some 
races are progressive in the evolutionary sense, while others are primitive. 
The two words, here simply Anglicized from the German, are apparently 
translations of Montandon's precoce and tardif. The distinction is that one 
is capable of further evolution, the other is not. In the von Eickstedt 
sense, the primitive branch is usually earlier. Thus he makes the Alpines, 
in particular, primitive; the Binaries, in contrast, are progressive forms 
of the same original root. 

According to von Eickstedt, the races which come under his classifi- 
cation entered Europe in post-glacial times. First came the Mediter- 
raneans, during the Mesolithic; then the Alpines, who approached the 
Swiss lake dwellings from the east, but still in Mesolithic times; the 
Binaries go back only to the Bronze Age. The Alpines were a forest 
people, and spread out into the forests of northern Europe as well as of 
those which covered the mountains in the center. An extra-primitive 
proto-Alpine type went to Benmark to associate itself with the Magle- 
mose culture. Then the Nordics broke through along the newly-formed 
northern steppes, and entered Scandinavia over Benmark, passing into 
Norway by two routes: around by Oslo; and through the gap between the 
two melting nuclei of the glacier, into Trondelagen. Earlier brachy- 
cephals are found at the termini of these routes. 

According to his system the Lapps are Alpines isolated in the north; 
they are the purest Alpines of all and are not mongoloid. The Nordics 

48 Nordenstreng, R., Europas Manniskoraser och Folkslag. 

Lundborg and Linders, Racial Characters of the Swedish Nation, pp. 5052. 
Hooton, E. A., Up from the Ape, pp. 508-5Q9, 535. 

49 Montandon, G., La Race, Les Races. 



are divided into three sub-divisions: Teuto-Nordic, the original and basic 
form; the Dalo-Nordic, which is the same plus Cr6-Magnon mixture; and 
a Fenno-Nordic, reddish haired and water-blue eyed, which is the eastern- 
most, largely Asiatic branch, now found only in solution. The Osteuropids 
are a separate race, a Nordic-Mongoloid transitional form, dating from 
the time of differentiation between these two stocks; and not a Nordic 
Mongoloid mixture, since its superior blondism and possession of dis- 
tinctive traits make its mixed derivation impossible. This race developed 
in the swamps and forests of the Obi drainage, and entered Europe only 
in modern times; its penetration of eastern and central Europe is a recent 

So much for von Eickstedf s classification. It fits with some fidelity the 
facts of racial distribution in Europe, but it does not fit all of the facts of 
history. In this respect we may apply the same criticism to the system of 
Czekanowski, which is illustrated by the diagram below: 50 

According to Czekanowski, there are four basic white races, located 
schematically at the corners of the square; and six sub-races or mixed 
types, which result from the crossing of the four fundamental ones. These 
races and sub-races, with their Greek letters, may be listed as follows: 















Pile Dwelling 


(alpha epsilon) 
(alpha lambda) 
(alpha chi) 
(epsilon chi) 
(epsilon lambda) 
(lambda chi) 

This scheme is obviously an attempt to place Deniker's system in a 
mathematically orderly form. Czekanowski defines his Lapponoid in 

Czekanowski, Jan, AAnz, vol. 5, 1928, pp. 335-359; AASF, ser. A, vol. 25, #2, 


such a way as to include the Alpine of Ripley, as well as the Lapps proper. 
In this identification of Lapps and Alpines, Czekanowski and von Eick- 
stedt agree. The Dinaric becomes a mixture of Lapponoid and Armenoid, 
which is difficult to follow; the "Pile Dwelling," being a mixture of Lap- 
ponoid and Mediterranean is, however, fully in accordance with the 
facts in regard to the crania of Swiss Lake Dwellers, 61 concerning which 
Gzekanowski is a specialized authority. 52 It seems unfortunate that the 
word "Alpine," should be torn from its context, immortalized by Ripley, 
and applied to a hypothetical Nordic- Armenoid cross, thus further abet- 
ting the confusion prevalent among even professional anthropologists, 
a confusion which Gunther, in his wholesale swapping of names, has done 
much to foster. 58 

It is not the purpose of the present survey to criticize in detail the two 
schemes chosen for presentation. Czekanowski, like Gunther, von Eick- 
stedt, and others, has rescued the Armenoid, which was first carefully 
described by von Luschan, 54 from the obscure companionship of Aus- 
tralians and Ethiopians in which Deniker had thrown it; he also, antici- 
pating von Eickstedt and following the early example of Pruner Bey, 55 
has attempted to salvage the Lapps from a mongoloid category and to 
make them full-fledged if primitive Europeans. But his scheme is mani- 
festly too pat, too regular, and too mathematical, to agree fully with 
nature, and, furthermore, it disagrees in many respects with the findings 
of the historical discipline. 

In making our own classification, let us first review the system which 
grew out of the skeletal study in Chapters II to VII. The groundwork of 
this system, and the list of types, may be gathered from the study of the 
lower half of Fig. 30. In this chart an attempt is made to separate the 
purely sapiens Mediterranean group from the Upper Palaeolithic mixed 
sapiens and Neanderthal races. Thus the Mediterranean sub-groups, 
races of food- producers which had already become differentiated before 
the great migrations into Europe, are listed as follows: Irano- Afghan, 
Corded, Atlanto-Mediterranean, Cappadocian, Mediterranean Proper, 
and Danubian. The old hunting and fishing population is divided into: 
Brunn, Borreby, and Alpine; while that branch which bears a considerable 
strain of incipient mongoloidism, includes Lappish and Ladogan, the 
latter being the vaguely mongoloid mixed meso- and brachycephalic 
element which appeared sporadically in the forest region of Russia, and 
occasionally to the south, from the beginning of the Russian Neolithic 

61 See Chapter IV, pp. 113-115. 

52 Czekanowski, J., AFA, vol. 48, 1925, pp. 65-76. 

88 Gunther, H., Rassenkunde der deutschen Votkes. 

** Luschan, F. von, JRAI, vol. 41, 1911, pp. 221-244. 

86 Pruner Bey, F., MSAP, vol. 2, 1865, pp. 417-432. 



onward. To this same side of the chart arc added the modern mongoloids 
and the mongoloid element in the American Indian. 

The lower half of the chart seems relatively simple in comparison with 
the upper portion, in which an attempt is made to show the relationships 
between these skeletal races and the living. The comparative simplicity 
of the lower portion, however, may reflect ignorance on our part rather 
than actual genetic isolation, since there was undoubtedly much mixing 
back and forth between the branches of each of the major lines, as well 
as between the lines themselves. 

The proposed classification of living whites and near-whites, which is 
shown on the top of the chart, may be listed in more detailed form as 


(1) Brunn: (Cr6-Magnon, to some extent) found in solution with Borreby, 
Nordic, and other elements, mostly in Scandinavia and the British Isles, 
also in North Africa and Canary Islands. May appear in comparatively 
pure form among individuals although nowhere as a total population. 

(2) Borreby; Large-headed brachycephals of Omet-Afalou type, the unreduced 
brachycephalic strain in Cro-Magnon; found in solution in peripheral re- 
gions of northwestern Europe, and as a major population element in most 
of northern and central Germany, and in Belgium. Like the. Briinn race, 
with which it is often associated, it occurs also in North Africa and the 
Canary Islands. 


(3) Alpine: A reduced and somewhat foetalized survivor of the Upper Palaeo- 
lithic population in Late Pleistocene France, highly brachycephalized; seems 
to represent in a large measure the bearer of the brachycephalic factor in 
Cr6-Magnon. Close approximations to this type appear also in the Balkans 
and in the highlands of western and central Asia, suggesting that its ancestral 
prototype was widespread in Late Pleistocene times. In modern races it 
sometimes appears in a relatively pure form, sometimes as an element in 
mixed brachycephalic populations of multiple origin. It may have served 
in both Pleistocene and modern times as a bearer of the tendency toward 
brachycephalization into various populations. 

(4)' Ladogan: I propose to give this name to the descendants of the mesocephalic 
and brachycephalic forest-dwelling population of northern Europe east of 
the Baltic in Kammkeramik times. This type is a blend of a partly mongo- 
loid brachycephalic element with a mesocephalic form of general Upper 
Palaeolithic aspect; these elements are seen in crania from Lake Ladoga 
and Salis Roje. (See Chapter IV, section 13, pp. 125-126.) Corded and /or 
Danubian elements are inextricably blended here, although the mongoloid 

66 Foetalization in a skeletal sense, which is, for obvious reasons, the only sense im- 
plicit here, involves a reduction of male secondary sex characters in the skull, and at 
the same time a reduction in skeletal sex differentiation. 


and Upper Palaeolithic elements seem at present more important. In its 
present form this composite type shows two numerous variants: 

(a) Neo-Danubian: Strongly mixed with the old Danubian, and to a lesser 
extent other elements, to form the common peasant type of eastern 
Europe, with many local variants. 

(b) East Baltic: Strongly mixed with Corded, Iron Age Nordic, and western 
Palaeolithic survivors to form the predominant population of much of 
Finland and the Baltic States. 

(5) Lappish: A stunted, highly brachycephalized, largely brunet relative of the 
Ladogan, originally living to the east of the Ladogan type area, in the Urals 
and western Siberia. Has probably assimilated some evolved mongoloid, 
but owes its partly mongoloid appearance more to the retention of an early 
intermediate evolutionary condition. In modern times much mixed with 
Ladogan and Nordic. 


(6) Mediterraneans: Within this general class, which still retains much of its 
original racial unity, the following sub-classes may at present be distin- 

(a) Mediterranean Proper: Short-statured, dolicho- and mesocephalic form 
found in Spain, Portugal, the western Mediterranean islands, and to 
some extent in North Africa, southern Italy, and other Mediterranean 
borderlands. Its purest present-day racial nucleus is without doubt 
Arabia. Most of the Cappadocian, isolated in the skeletal material, 
seems to have been absorbed into the western Mediterranean variety 
after its early Metal Age migration, while that which remained in Asia 
Minor became assimilated into the Dinaric and Armenoid. It still ap- 
pears, however, among individuals in its original form, and is particu- 
larly common among Oriental Jews. 

(b) Atlanto-Mediterranean: The tall, straight-nosed Mediterranean, not meso- 
cephalic, as Deniker erroneously stated, but strongly dolichocephalic. 
Today this race forms the principal element in the population of North 
Africa, and is strong in Iraq, Palestine, parts of Arabia, and the eastern 
Balkans; in solution with varying degrees of negroid it is also the prin- 
cipal race in the whole of East Africa. In Europe it is a minority ele- 
ment in the Iberian Peninsula, Italy, and the British Isles. 

(c) Irano-Afghan: The long-faced, high-headed, hook-nosed type, usually of 
tall stature, which forms the principal element in the population of 
Iran, Afghanistan, and the Turkoman country, and which is also present 

- in Palestine, parts of Arabia, and North Africa. It is probably related 

to the old Corded type of the Neolithic and Bronze Age. 
(7) Nordics: The basic Nordic is the Corded-Danubian blend of the Aunjetitz 
and of the Early Iron Age in central Europe. This type includes some Bell 
Beaker Dinaric absorbed in early Metal Age times. Although Danubian 
and Corded types may appear as individuals, they may nowhere be isolated 
as populations. The most important living Nordic varieties are: 
(a) Keltic Iron Age Type: The Keltic sub-type, mesocephalic and low-vaulted, 
with a prominent nose. Commonest in the British Isles where in places 


it forms the principal element in the population, Also a major element 
in Flanders and the Prankish country in southwestern Germany. 

(b) Anglo-Saxon Type: The old Germanic Reihengraber type, a heavy-boned, 
rather high-headed Nordic variety, most prevalent in northern Germany 
and England. 

(c) Trondelagen Type: A hybrid type of Nordic with Corded and Briinn 
elements, frequent in the central coastal provinces of Norway, north 
of the Dovre Mountains; the principal form in Iceland, and among 
the Frisians, and common in the British Isles, The Anglo-Saxon type 
lies between it and the true Nordic. 

(d) Osterdal Type: The original Hallstatt Nordic, smaller-headed and finer 
boned than (b) or (c); occurs in many populations as individuals, 
typical only in Sweden and in the eastern valleys of Norway. 


(8) Dinarics: A tall brachycephalic type of intermediate pigmentation, usually 
planoccipital, and showing the facial and nasal prominence of Near Eastern 
peoples. The basic population of the whole Dinaric-Alpine highlands from 
Switzerland to Epirus, also in the Carpathians and Caucasus, as well as 
Syria and Asia Minor. Apparently a brachycephalized blend in which 
Atlanto-Mediterranean and Cappadocian strains are important, with Alpine 
acting as the brachycephalizing agent in mixture. Borreby and Corded 
elements, also Nordic, appear to be involved in some regions. 

(9) Armenolds: A similar brachycephalic composite type, with the same head 
form as the Dinaric, but a larger face and nose. The pigmentation is almost 
entirely brunet, the pilous development of beard and body abundant, the 
nose high rooted, convex, and the tip depressed, especially in advanced age. 
The difference between the Armenoid and the Dinaric is that here it is 
the Irano-Afghan race which furnishes the Mediterranean element, brachy- 
cephalized by Alpine mixture. 

(10) None: A blond, planoccipital brachycephal frequently encountered in South 
Germany and elsewhere in central Europe. This is apparently an Iron Age 
Nordic brachycephalized by Dinaric mixture and seems in most respects to 
take the form of a blond Dinaric variant. Both Deniker and Czekanowski 
have recognized this type, and it is a standard race, under various names, in 
most Russian studies. The name Noric was given it by Lebzelter. A brachy- 
cephalized Neo-Danubian, common in Jugoslavia, is a parallel or variant 

The ten racial types within the white race listed above, with their sub- 
types, form two of the three main divisions of the white race, in its widest 
sense, when segregated on the basis of head size. The third division, that 
of the peoples with small heads, includes the aboriginal population of 
southern Arabia east of the Yemen, and various groups in Baluchistan, 
and again in southern India. This third variety is characterized by an 
abundance of wavy or ringleted hair, and facial features of a Veddoid 
character which in some instances suggest Australoid affinities. This third 


o o o o o o.o 

O O o'l O O O O O O O O 

o o o o o o o o o oooooo 






strain, Tronder, etc., yn 
d,only partly bradiycepna 


rvivor (Teviec Type, 


Neo-Danubian element as 



A A 
A A A A A A 

AAAAA .. . 
A. A A A A A A 
A A A A A A 

A - -v*-A,/ 

A A - - 



A A 'A~\A A A A A A A 


division need not, however, concern us here, because it falls outside the 
major range of the white race. It will be dealt with in some detail in the 
proper section of the regional study. 

Besides the European races proper, as listed in the preceding para- 
graphs, and' their Veddoid collaterals, there are certain fully evolved non- 
white races which have influenced the European population by intrusion 
and blending. These include at least two of the sub-divisions of the mon- 
goloid family the Buryat- Mongol, to which the Avars in part belonged, 
and which is today represented on European soil by the Samoyeds; and 
the Tungusic, the type of the early Huns. To these may be added an 
apparently stabilized mixed form, resembling a partially mongoloid 
Dinaric, to which many central Asiatic Turkish tribesmen belong. In 
addition to these Asiatics, there remains the African Negro, which has had 
certain influences upon the formation of race in the Mediterranean region, 
especially in North Africa, and in parts of Arabia. Other non-white 
stocks, such as the Australoid, Negrito, and Khoi-San (Bushman-Hotten- 
tot), have not affected the white group in its homelands in any discernible 

Chapter IX 


The remaining chapters of this book will be devoted to a rapid survey 
of the continent of Europe, country by country and people by people, and 
of the contiguous portions of Asia and Africa occupied by basically white 
populations. The treatment of the skeletal documents in prehistory and 
history, and the survey of the living material as a whole, which have pre- 
ceded this section, will make elaborate introductions unnecessary. Here 
it is proposed to cover the geography of the white race piecemeal, for the 
convenience of the reader interested in specific local problems, as well as 
to examine in further detail the nature of the white human division as a 

Every map is two dimensional, and every consecutive written work one 
dimensional. There is a conflict, therefore, at the start between the nature 
of any geographical material and the medium through which it is to be 
described and explained. The choice of a starting point is a purely arbi- 
trary affair, and the sequence of areas followed must be equally dogmatic. 
Perhaps because of our European habit of starting a written page at the 
upper left hand corner and working down, strip by strip, we shall follow 
this system, more or less, in our study of the map of Europe. 

By following this method we shall first deal with the very northernmost 
zone, which is, in effect, a more or less unified environmental area. It is 
at the same time the last portion of the European land-mass to receive 
permanent settlement, and the last to receive the cultural stimulus 
of agriculture. For these and other reasons, all of which resolve them- 
selves ultimately into the fact that northwestern Europe was the center 
of Old World glacial activity during the last age of ice, the far north has 
played the zoological r61e of a marginal area. Its racial history, while 
complex enough in the absolute sense, is relatively simple and relatively 
easy to untangle, as has been shown in previous chapters. 

Aside, from the Russian Slavs whose appearance in the north is of 
recent historical date, ,we have, in this zone, to deal with two linguistic 
groups the Uralic, with sub-divisions into Finnic, Ugric, and Samoyedic; 
and the Indo-European, in Scandinavian and Baltic forms. From the 
standpoint of race in the sense of major world groupings, we are con- 
cerned with two the white and the mongoloid. In the historical sense, 



we are confronted again with a division between Palaeolithic survivors, 
and the descendants of the farthest wandering of Mediterranean food- 
producers. From the standpoint of environmental conditioning in its 
effect upon the human form, we have reached an area of maximum 
differentiation. Northern Europe, especially northwestern Europe, has 
served not only as a refuge area for archaic humanity, but also as a source 
from which migrations of vast compass have spread southward into 
warmer lands at times of environmental distress. Emigrants forced out 
by the vagaries of its treacherous climate have not only affected in varying 
measure the rest of Europe, but have likewise played a principal part in 
the peopling of the New World. 


If the white race spreads far beyond the arbitrary boundaries of the 
European continent to the south and east, the opposite may be said of 
the north. In the circumpolar zone which fringes the Arctic Sea, Asia 
encroaches upon Europe, and except for Iceland, the racial uniformity 
of this frigid ring is, superficially at least, complete. In a far less super- 
ficial sense, is the cultural uniformity valid. From Greenland to Lapland 
one finds short, lank-haired people driving across the frozen tundra in 
bone-shod sleds, drawn by dog or reindeer; these hyperboreans dress 
themselves in warmly tailored fur garments, with trousers for both sexes 
alike; they live in conical huts of birch bark, or domes of rock and sod; 
they venerate the bear and witness the supernatural spirit flights and 
ventriloquistic conversations of their shamans. 

With few exceptions they are all short in stature, and this shortness 
reaches its extreme at the two ends of the circumpolar zone, Greenland 
and Lappland. This shortness is accentuated in all of the circumpolar 
groups by a relative reduction in leg length, with a greater trunk height. 
The same reduction in length, probably produced by the same mechanism, 
has been noted in the case of the Magdalenian hunters in late glacial 
times. These same Magdalenians, notably Chancelade and the male 
Obercassel, showed at the same time an incipient degree of mongoloid 
adaptation, insofar as this adaptation is visible in the skull and especially 
the facial skeleton. It is likely that the occurrence of partial mongoloid 
traits in many Upper Palaeolithic survivor groups may be due to the 
retention of traits acquired during the final glacial maximum. In the 
same way all of the circumpolar groups show, in one degree or another, 
a certain amount of mongoloidism, and it is possible that the mongoloid 
stock as a whole represents a progressive mutation from a proto-white 
stock, of Upper Palaeolithic variety, which began in the Late Pleistocene 
and reached various degrees of specialization in post-glacial times. 


The westernmost representatives of this circumpolar ring of peoples are 
the Lapps, who call themselves, in their own archaic variety of Finnic 
speech, Samen. Their country, Lapland, has no political existence, but 
is no less real an entity. It consists of the forested highlands of northern 
Sweden, which afford ideal reindeer pasturage, and the tundra-covered 
stretches of northern Finland, with the Norwegian coastal provinces of 
Troms and Finnmark, and much of the Russian Kola Peninsula. Except 
for small patches of forest and mountain, the Lapps are not alone in this 
country, but share it with a more numerous population of Finns and 
Norwegians, with whom they have, for centuries, been mixing. 

There are, in the whole world, probably no more than 32,000 Lapps. 1 
Of these about 21,000 live in Norway, 7000 odd in Sweden, and 3000 
more are evenly divided between Finland and Russia. In Norway, which 
holds thus two- thirds of the total, between ten and eleven thousand are 
concentrated in the province of Finnmark, where, in 1920, they formed 
24 per cent of the population. In Sweden the greatest concentration is in 
Norrbottens Ian, which holds 4500. The Lapps are not, from the stand- 
point of numbers, an important people in the world. They are one of 
the marginal, vestigial groups destined to disappear by the process of 
absorption. Their importance lies, however, in their taxonomic position, 
and in the influence which they have had in the past, and may have in 
the future, on other European peoples with whom they have blended and 
will blend. 

Their predilection for this blending process is so great that it is really 
very difficult to estimate their numbers, and the figures given above are by 
no means definitive. They include Lapps who speak their own language 
and call themselves ethnically Samen, and exclude those who have passed 
over into other populations, notably the northern Norwegian. At the 
same time they include many Norwegian, Swedish, and Finnish genetic 
lines which have been incorporated intcf the culturally Lappish body. 

Norwegian writers usually divide the Lapps into two main classes, the 
Reindeer Lapps, living in the forests and mountains, and the Sedentary 
Lapps, living along the coast and rivers, subsisting mostly on fish. It is 
generally believed that the original Lapps who entered Scandinavia were 
reindeer-herders, and that for many of them the sedentary life is a rela- 
tively recent readaptation. Today, however, no more than five thousand 
still herd reindeer, and of these five, three live in Sweden. Thus although 
Norway holds the majority of the world's Lapps, those who preserve the 
purest Lappish type, both in culture and race, live over the Swedish border. 

The Lapps present a distinct problem to students of race, which has 
been answered in one way or the other by various authors since the middle 

* Wiklund, K. B., GE, vol. 13, 1923, pp. 223-242. 


of the last century. The problem is: are they primitive European brachy- 
cephals, related to the Alpines of west central Europe, or are they mon- 
goloid invaders from Asia? This question is of more than taxonomic 
value, because it is intimately concerned with the historical position of 
all the western European brachycephals as well as with the validity of the 
classifications employed by the present schools in Poland and Germany. 
Fortunately, with the publication in 1935 of Schreiner's %ur Osteologie der 
Lappen^ we are at length in a position to answer the Lapp question in a 
definite manner, and with some degree of assurance. The answer lies 
partly in the historical field, and partly in that of somatology. 

The historical evidence does not favor the Alpine or local shrunken- 
Palaeolithic-survivor theory. In the first place, the Lapps speak a Finnic 
dialect which is classified with the extinct Chude, spoken in the early 
centuries of the present era in Finland and the regions immediately east 
and north of the present city of Leningrad. 3 The Chudes were Volga Finns 
who migrated in early times into the regions later to be occupied by their 
modern Finnish and Esthonian relatives, who eventually absorbed them. 
In the Lappish language are also found certain loan words from Letto- 
Lithuanian, and others from early Scandinavian. Letts and Lithuanians 
arrived in the Baltic lands only in the middle of the first millennium A.D. 
Then the Lapps could not have moved to the far northwest much before 
this time. Furthermore, in order to have borrowed their language from 
the Ghudes, who themselves did not arrive there much earlier, the Lapps 
must have mixed to some extent with them, and indeed the Lappish 
skeletons disinterred in Scandinavia show mixture with a Finnic type 
from the beginning. 4 

In the fourteenth century the Lapps were mentioned in the Lake Onega 
region, and tax registers from the sixteenth century establish their pres- 
ence as far south as Lake Saima, a short distance farther north; hence it 
is certain that the Lapps had not been fully pushed up into their Arctic 
environment until recent times. In Norway, the earliest graves, found in 
Finnmark, may date from late "Roman" times, near the middle of the 
first millennium A.D., but the presence of the Lapps in this country is not 
absolutely certain before the ninth century. At this time Norse traders 
and settlers were sailing around the North Cape, into the previously un- 
known provinces of Troms and Finnmark, and they met Lapps there and 
mixed with them. A rich Viking grave of the tenth century, in eastern 
Finnmark, contains the skeleton of a twenty year old youth of manifestly 
mixed Norse and Lappish ancestry. 5 

2 Schreiner, K. E., %ur Osteologie der Lappen. 

8 Wiklund, K. B., loc. cit. 

4 Schreiner, K. E., op. cit., vol. 2, p. 279. 

6 Schreiner, Alette, Antkropologische Untersuchungen in Norge; Hellemo. 


Schreiner has collected some 300 Lapp skeletons from graves along the 
Norwegian coast, all of which were of Lappish construction or contained 
typically Lappish grave furniture; there is no reason to confuse them 
either with contemporary Viking graves or with the earlier remains of 
the Stone Age people of this region, for the Lapp graves are manifestly 
late and intrusive. Furthermore they are geographically restricted, for 
the Lapps did not, before the sixteenth century, range below 63 N. Lati- 
tude, and the most southerly Lappish burials yet found are at Steinkjaer 
on the inner Trondhjem fjord. The Lapp inroad of the eighteenth century 
into South Trondelag and Hedmark came not from the north, but from 
the Swedish provinces of Jamtland and Hardjedalen, to the east. The 
Lapps did not, therefore, extend south into central Norway until very 
recent times, and had no opportunity to mix with Norwegians in any 
numbers south of Tydsfjord, the northernmost fjord-valley of Nordland. 
They cannot, therefore, have been responsible for the brachycephaly in 
southern Norway. Although there is no skeletal material from the Stone 
Age sites of northern Norway, there is no reason to suppose that these 
people were the ancestors of the Lapps, since Lapp sites and Stone Age 
sites are distinct, and nothing transitional has been found. 

On the historical side, the evidence is clear. In regard to somatology 
we may be equally positive, since there is no lack of anthropometric 
material. Series by Bryn, 6 Alette Schreiner, 7 Gjessing, 8 Geyer, 9 Kajava, 10 
and Zolotarev n represent Lapps from Norway, Sweden, Finland, and 
Russia; these studies are all modern and cover the living material fully, 
while K. E. Schreiner's skeletal series provides a check upon the dead. 
All of these series show that the Lapps are very mixed, and that they 
contain not only Nordic blood, derived from Norwegian contact, in- 
tense during the last four centuries, but also a blond brachycephalic 
element which presumably comes from their even commoner mixture 
with the Kvaens, the northernmost of Finns. Several attempts have 
been made to isolate "pure" Lapps, but this isolation must be relative 
since they were probably mixed before they arrived in the present Lapp- 

It is generally assumed that the Lapps were originally brunet, and that 
what blondism they possess has been acquired through this mixture. 
There is, of course, no factual basis for this assumption, and if it be true, 
the Lapps must have more non-Lapp than Lapp blood. Hair color was 

Bryn, H., MAGW, vol. 62, 1932, pp. 1-74. 

7 Schreiner, A., Die Nord-Norweger; Hellemo (Tysfjord Lappen). 

8 Gjessing, R., Die Kautokeinolappen. 

9 Geyer, E., MAGW, vol. 62, 1932, pp. 163-209. 

10 Kajava, Y., Beitrage zwr Kenntnis der Rasseneigenschaften der Lappen Finnlands. 

11 Zolotarev, D. A., Kolskie Lopari. 


observed by means of the Fischer scale in six modern studies, 12 while in 
two others l3 no scale was employed, but the material is capable of com- 
parative use. In these series the adult male Lapps vary, in black to dark 
brown hair colors, from forty to eighty per cent; the beard color, when 
observed, is lighter. There is some argument as to whether the pure 
brunet Lapp hair is really black or dark brown, which would indicate 
that it often falls into a borderline category. When blond, it is usually 
ashen, and almost never golden or red. The selected "pure" groups, 
Bryn's Reindeer Lapps, and some of Geyer's mountain and forest Lapps 
from Sweden, have seventy per cent or over of this dark hair, while the 
fairest Lapps, with a majority of brown and blond shades, are found in 
Finland and in the Kola Peninsula. 

Pure dark eyes are found among one-third of Reindeer Lapps, and 
among as few as eight per cent in the total of Lapps from Norway. 14 Pure 
light and light-mixed eyes are commonest among the Lapps of Finland, 
where they total between thirty and forty per cent, and least common 
among the Reindeer Lapps of interior Norway and Sweden. Even among 
the purest selected sub-groups, such as that of Geyer, who isolated from a 
larger Swedish Lapp sample a few individuals of most pronounced Lap- 
pish type, at least a third are light or light-mixed in iris color. 

The skin color of Lapps with light hair and eyes is as light as that of 
Norwegians and Finns, but in the majority, with mixed or dark hair and 
eye pigmentation, the skin tends to a grayish yellow to yellowish brown, 
with some moderately dark individuals, equivalent in pigment intensity 
to Spaniards or Italians. 15 On the whole the skin is lighter on the face 
and darker on the body, and is usually darkest on the abdomen and 
genitalia. 16 Among the old this skin becomes deeply wrinkled, since it is 
then deficient in sub-cutaneous fat. The eyes are set in deeply excavated 
sockets in senility, owing to the same fat deficiency. 

The Lapp hair is thick on the head, usually straight or but slightly 
wavy; it is of moderate texture, and seldom coarse or wiry in a truly 
mongoloid manner. Graying begins late, and baldness is rare. The beard, 

12 Bryn, H., MAGW, 1932. 
Geyer, E., MAGW, 1932. 
Gjessing, R., Die Kautokeinolapptn. 
Schreiner, A., Die Nord-Norweger; Hellemo. 

Luther, M., unpublished data in Peabody Museum. Actual hair samples collected, 
and later matched in the laboratory. 

13 Kajava, Y., Beitrage %ur Kenntnis der Rassemigenschaftm der Lappen Finnlands. 
Zolotarev, D. A., Kolskie Lopari. 

14 Schreiner, A., Die Nord-Norweger, Martin's numbers 2-4, total of 254 males. 

16 Bryn, H., MAGW, 1932, finds 20 per cent to have von Luschan #3; the darkest 
shade which he records is #12. 
16 Schreiner, A., Hellemo^ p. 15. 


except where much Nordic blood is apparent, is very scanty, consisting 
of a few widely separated hairs. The body hair again is largely deficient, 
for there is seldom any on the chest or abdomen; even the pubic hair is 
scanty, and on men as well as on women its area of growth is sharply out- 
lined at its upper border. The external genitalia which this hair partly 
conceals are as a rule small. 

Thirty different series give the stature of the Lapps over a span of 
130 years. Eleven series published between 1870 and 1900, give means 
of 138 to 156 cm., which could be averaged at 151 cm. Twenty-seven 
others, measured between 1905 and 1934, and including several thousand 
Lapps, range from 155 to 164 cm.; 17 during this period the Lapps grew, 
apparently, seven or eight centimeters. This may be accounted for either 
as evidence of continuous progressive mixture, or the influence of the 
stature increase tendency in northwestern Europe, or both. 

A study of Lapp bodily proportions shows that the trunk is long in 
proportion to the legs, which are especially short in the tibial segment, 
and often bowed; the arms are relatively long, especially in the humeral 
segment. The hands and feet are as a rule small and delicate. Despite 
the great relative arm length, both shoulders and hips are narrow, and 
these peculiarities are especially accentuated in the more brunet, shorter, 
and presumably less mixed segment of the Lapp population. 18 

The head of the Lapps, while large in proportion to the body size, is 
absolutely small. The length ranges in the low 180's, and the breadth 
in the 150's, while the height is probably about 122 mm. The cephalic 
index means range from 80 to 88; and a large list of series shows no change 
during the last century. There are, however, regional differences; the 
center of extreme round headedness lies among the inland groups in 
northern Norway, while the Swedish, Finnish, and Kola Peninsula Lapps 
become progressively narrower headed. The mean for the purest Rein- 
deer Lapps of Norway is 87; for the easternmost Lapps, 80 to 83. 

The forehead of the Lapps is narrow in proportion to the parietal 
breadth; the profile of the head from above is a short ovoid. The occiput 
is flat-curved, with some flattening at lambda. Browridges, as a rule, are 
absent, and the forehead is usually steep, and frequently equipped with 
frontal bosses. The faces of the Lapps are extraordinarily short, with well- 
substantiated means of 112 mm. for nasion-menton height. These may be 
compared with the means of 124-126 mm. usual among either Norwegians 
or Finns. In this the Lapps differ from known whites or mongoloids to 
an extraordinary degree, and an extreme absolute facial shortness must 

17 For a complete bibliography of early Lappish series, see the lists of Bryn, the two 
Schreiners, Geyer, Kajava, and Zolotarev. 
8 Geyer, MAGW, 1932. 


be considered a distinctive Lapp feature. Upon further examination, it 
may be seen that this shortness lies almost entirely in the masticatory 
segment of the face height; the alveolar borders of the maxillae are extra- 
ordinarily shallow, and the mandible is very low, weak, and feebly de- 
veloped. 19 The jaw is not, however, narrow at the rear, for the bigonial 
diameter is as great as 108 mm. on Norwegian Lapps, and greater even 
among "pure" Nomads. 20 The jaw converges rapidly toward the chin, 
which is small, pointed, and frequently receding. The teeth are very small, 
and their roots short. Thus the Lapp face is distinguished by a reduction 
of jaw size and an oral shallowness extreme and perhaps unique among 
mankind. It must be considered as a Lapp specialization coincident with 
their extremely short stature, and especially with the shortening of the 
distal leg segment. 

Otherwise the Lapp face takes a position midway, in many respects, 
between whites and rnongoloids. The bizygomatic diameter, of 140 mm., 
or thereabouts, is in the white range; it is narrow in proportion to the 
vault, but it seems wide in relationship to jaw and forehead. The malars, 
while not notable for lateral jut, project forward prominently. The nose 
is on the whole low and flattish; with a straight or concave bridge, low 
root, and a peculiar snubbed or pointed, up-turned tip. This prominence 
of the tip is retained characteristically in mixture. On the whole the nose 
is mesorrhine, and is in this respect not unlike those of many of the 
Finnish and Slavic peoples in eastern Europe. The eyes are widely 
separated, set in low orbits, and overhung in some instances with median 
or external folds, rarely with the mongoloid epicanthus. 

On the whole, the Lapp crania, as the Lapp soft parts, take an inter- 
mediate position between mongoloid and white standard forms. In some 
special characters the Lapps are unique, as in the masticatory develop- 
ment, and in the orbit, where Hisinger-Jagerskiold has found a curiously 
primitive bony conformation. 21 The possession of these peculiar special- 
izations and primitive traits should prevent the Lapps from being con- 
sidered a hybrid mongoloid-white racial form. Compared to central 
Asiatic mongoloids, the Lapps are little specialized. The soft and often 
fine head hair, the absence of the blue-black hair pigment shade, the 
infrequency of the mongoloid eyefold, and the absence of an excessive 
lateral malar development or of great facial width, are evidence of this 
lack of specialization in a mongoloid direction. 

There are many features which give the Lapps an infantile appearance 
which cannot be accidental; these include the body and limb proportions, 

19 Schreiner, K. E,, ur Osteologie der Lappen. 
2 Bryn, H., MAGW, 1932. 

21 Kajava, Y., Beitrdge zur Kenntnis der Rasseneigenxchaftm der Lappen Finnlands, p. 35, 
after Hisinger-Jagerskiold, E., FFVS, vol. 55, 1913. 


the sparseness of body hair, the small genitals, the bulbous forehead with a 
smooth supraorbital region, the weak chin, and the low, child-like nose. 22 
Some environmental mechanism working upon the mineral economy of 
this peripheral human group has probably produced this size reduction 
and infantilism. 23 

Schreiner's opinion, based upon a detailed study of Lapp craniology 
as well as upon the living material, is simple and adequate. Translated 
into the terms of the present study, it signifies that the original ancestral 
Lapps represented a stage in the evolution of both the Upper Palaeolithic 
Europeans and the mongoloids, and that while the mongoloids have 
specialized in their own characteristic way, and while the Ice-Age Euro- 
pean strain was modified by mixture with and virtual absorption by the 
encroaching post-Pleistocene food producers, the ancestral Lapps were, 
in their turn, modified largely by a general size reduction and an increas- 
ing infantilism. The jaw reduction of the Lapps is their most easily iden- 
tified specialization. 

In view of the known history of Upper Palaeolithic whites and of mon- 
goloids, this divergence of the Lapps from the others must have taken 
place as early as the Laufen glacial retreat. Their area of specialization 
was presumably western Siberia, where they found room in which to 
specialize with little interference. From here they must later have spread 
over Finland and northwestern Russia, whence they entered northern 
Scandinavia sometime during the first millennium of the Christian era, by 
a gradual trickling process. In their northern wanderings they may have 
met the Samoyed, and from them acquired their domestic reindeer and 
the habit of reindeer milking. Since, according to both Laufer and Hatt, 24 
this last trait did not develop in its central Asiatic home much before the 
middle of the first millennium B.C., the Lapps could not have acquired this 
practice much before their arrival in Scandinavia. The acquisition of this 
superior economy must have given them an impetus for northward ex- 
pansion, as it did, farther east, with the'Tungus. 

We must not look for Lappish ancestors, therefore, in the large-headed 
Borreby people of Mesolithic and Neolithic Denmark, nor in the occu- 
pants of the Stone Age sites of northernmost Norway; if we find Lapp-like 

22 This general estimate of the Lapp racial position is for the most part a paraphrase 
of K. E. Schreiner's conclusions in his %ur Osteologie der Lappen, by far the most erudite 
work yet to appear on the Lapp question. 

23 Gjessing, R., Die Kautokeinolappen, pp. 90-95. 
Marett, J. R. de la H., Race, Sex, and Environment. 

24 Hatt, G., Notes on Reindeer Nomadism, MAAA, vol. 6, 1919. This is one of the few 
points regarding the history of reindeer husbandry upon which these two authorities 

Laufer, B., The Reindeer and Us Domestication, MAAA, vol. 4, #2, 1917; AA, vol. 22, 
1920, pp. 192-197. 


physical traits, as do Czekanowski, Mydlarski, and others, among eastern 
European brachycephals, and even among western European Alpines, 
we must remember that some of the Lappish peculiarities, including 
perhaps their specialized nasal tip form, may have been common posses- 
sions of the Upper Palaeolithic European peoples as well. As we shall see 
later, there may have been transitional forms between Lapps and Euro- 
peans, and this general class of humanity may be responsible for the wide- 
eyed brachycephals who, as we saw in our historical chapters, appeared 
now and then in southern Russia and Poland from the beginning of the 
Neolithic onward. 


In the eastern extension of their territory the Lapps share the Kola 
Peninsula with their neighbors and fellow reindeer-herders, the Samoyeds. 
The Lapps represent, however, a much older population, for the Samoyeds 
have only lived there for a few centuries. There are, according to Russian 
authorities, only V>,500 Samoyeds in the world; of these 4000 live on the 
Kola Peninsula, another 5000 range between the White Sea and the 
mouth of the Yenisei River, and the rest hunt between the Obi and 
Yenisei rivers, and in the Yenisei drainage. Thus the bulk of the Sam- 
oyeds still inhabit their Siberian home. All of those mentioned speak a 
language which constitutes one of the two primary divisions of Uralic 
speech. It seems to be definitely related to Finno-Ugrian, although its 
supposed kinship to Tungus, Mongol, and Turkish has been questioned. 

Other Samoyeds, who have been Turkicized in language, and to a large 
extent in manner of living, dwell in southern Siberia, in the provinces of 
Yeniseisk, Tomsk, and Irkutsk, and also in Mongolia. These go under 
the names of Soyots, Karagas, and Uriankhai; they are more numerous 
than the Samoyeds proper. Whatever their earlier history, the Samoyeds, 
without reasonable doubt, may be considered to have developed as an 
ethnic and linguistic group in the region north of the Altai Mountains, 
the general center of Altaic-speaking Mongoloids. 26 Their spread north- 
ward into Siberia and thence to the Arctic rim of Europe must have been 
a relatively recent phenomenon. 

In . central Asia the Turkicized Samoyeds are definitely and fully 
mongoloid, and belong to the Buryat-Mongol variety, which we have 
encountered historically among the Avars. Those who live in Europe have 
brought the same physical type with them with but little modification. 

26 Non-anthropometric data mostly from Jochelson, W., Peoples of Asiatic Russia; and 
from Les Voyages du Professeur Pallas. 

26 Professor G. J. Ramstedt of Helsingfors University has expressed the opinion that 
the original bearers of Samoyedic speech must at one time have moved to the Altai 
region from a point nearer the Finno-Ugrian homeland. Private Communication. 


Cranially they resemble the Lapps in vault size and dimensions, 27 but the 
Samoyed facial skeleton is wider and larger, with a more nearly mon- 
goloid development of the malars. 

Our material on the living is sufficient in numbers and detail to permit 
a confirmation of this mongoloid character. 28 The stature, with a mean 
of 154 cm. in 1887, had risen to 156.8 cm. in 1914, providing that the 
same groups were represented. Like the Lapps, the Samoyeds are short, 
and like them relatively long bodied. They are brachycephalic, but not 
to the extent attained by the western Lapps in Norway; they are eury- 
prosopic and mesorrhine. 

Although the mongoloid character of the Samoyed may easily be seen 
in their flattish, round faces, everted lips, and up-tilted, low-bridged nose, 
and in their scarcity of beard, one cannot call them purely or completely 
mongoloid. Photographs of Samoyeds 29 show a considerable number 
with partially European features. Sommier's data on hair and eye color, 
again, shows some thirty per cent with mixed or light eyes, and the same 
number with hair ranging from medium brown to blond. As with the 
Lapps, the women are notably darker in hair and eye color than the men. 
This pigmentation variability, in view of the sex linkage, would indicate 
that the Samoyeds as well as the Lapps, but in lesser degree, had been 
subjected to mixture with peoples of European racial character. This 
mixture may be explained in several ways: (a) by the retention of an 
early non-mongoloid condition derived from ancient Uralic-speaking 
ancestors; (b) by contact with central Asiatic Nordics before the depar- 
ture of the Samoyeds for Europe; (c) by mixture with Ugrians, Finns, 
Slavs, and others in western Siberia and northern Russia. 


The northern zone of Europe which we have chosen as the subject of 
our first regional chapter is in reality two zones; besides the northern- 
most, which runs closely around the Arctic rim, is a second, that of Scan- 
dinavia and the lands to the east of the Baltic. Here food production is 
possible, and the effect of environment does not necessarily take the form 
of infantilism and stunting. In contrast to the first, this second zone is 
occupied by large, typically European groups of people. 

27 Schreiner, K. E., %ur Osteologie der Lappen, pp. 280-281. 

Sommier, S., APA, vol. 17, 1887, pp. 71-222. 

Klimek, S., ibid., vol. 59, 1929, pp. 13-31. 

^Roudenko, S. I., BMSA, set. 6, vol. 5; 1914, pp. 123-143. 

Sommier, loc. cit. 

Zograf, N, J., AAM, 1879, vol. 2, pp. 61-87. From resum6 by Stieda, AFA, vol. 14, 
1883, p. 291. 

29 Peabody Museum Collection. Courtesy of the Institute of Northern Peoples, 


Norway occupies the poorer and more rugged half of the Scandinavian 
Peninsula. The mountain crest which separates it from Sweden runs to 
the west of a central line, and swings to the northeast in such a way as to 
give to Norway the northernmost part; so that much of Norway, and rela- 
tively little of Sweden, lies within the Arctic circle. Deep fjords along 
most of the Norwegian coast cut far into the land, in some cases nearly 
bisecting the kingdom. A large proportion of the country is mountainous, 
but aside from the central spine, only one range deserves mention here 
that of the Dovre Mountains, which separates M0re and Tr^ndelagen 
on the north from Opland and Hedmark on the south. Only in the long, 
eastern valleys such as 0sterdal and Gudbrandsdal, and on the plain of 
Oslofjord, are large unbroken stretches of reasonably flat farm lands to 
be found. 

The topography of Norway, as outlined above, is important in its effect 
upon the present distribution of its peoples. While Sweden, a lake- 
studded plain sloping gently from the western mountain barrier to the 
Baltic, is inhabited by a regionally uniform population, Norway, with its 
rugged fjords and deeply folded valleys, provides shelter and differentia- 
tion room to a number of local types. Norway's geography, in combina- 
tion with her climatic and cultural history, makes her one of the most 
marginal areas, in a racial sense, in Europe. Yet, despite her marginal 
character, Norway has played an important part in European racial his- 
tory, since this nation has been a source of emigration to Iceland, to 
Normandy, and to the British Isles. Hence it has had much to do with 
the modern settlement of the New World, which Norwegians discovered. 
The physical types of many British and Americans may be traced directly 
to a Norwegian origin. 

The racial history of Norway has been covered, insofar as we know it, 
in the preceding chapters. The northern coastal regions had a very late 
age of chipped stone, and an even later Neolithic. Food-producing peoples 
were few in Norway before the Middle and Late Bronze Ages, and 
not until the Iron Age and the full Viking period was this country fully 
inhabited. The greatest pre-Iron Age concentration was along the south- 
western coast. Nordland, Troms, and Finnmark were explored during 
the pre-Christian Viking period; Troms and Finnmark were abandoned 
during the Middle Ages and only resettled from the sixteenth century 
onward. Our skeletal material, wholly Iron Age in date, shows a medley 
of normal Iron Age Nordic crania with Borreby brachycephals and other 
skulls which could be fitted without difficulty into a dolichocephalic 
Upper Palaeolithic category. That the Norwegian climate has not 
exerted a size-reducing tendency such as that which somewhere modi- 
fied the ancestors of Lapps, is therefore shown by the survival of 


what appear to be full-sized Ice-Age cranial types into the present millen- 

The study of living Norwegians has been carried on with excep- 
tional competence by three modern investigators Halfdan Bryn, K. E. 
Schreiner, and Mme. Alette Schreiner. The monumental Somatologie der 
Norweger^ by the first two named, supplemented by the regional studies 
of the first and third, 31 continue the earlier work of Arbo, Helland, Larsen, 
and the Daaes, 32 and present us with a body of accurate and objectively 
interpreted anthropometric data unsurpassed elsewhere. 

The Norwegians, as a whole, are tall by absolute standards, and blond, 
with moderate body proportions which include relatively long legs and 
short arms. Most of them are mesocephalic, with meso- to leptoprosopic 
faces, and their noses are usually leptorrhine. Regional variations in 
Norway are relatively great for Scandinavia, but are no greater than those 
found in many European countries. Except for the far North, local 
stature means run from about 168 to 175 cm., while the cephalic index 
varies by parishes between the extreme means of 76 and 84. In hair and 
eye color, blonds and mixed forms are everywhere more numerous than 
brunets; dark eyes, for example, never reach the figure of 20 per cent. 

Within these relatively restricted ranges there is a definite pattern of 
regional distribution, and there are four definite areas, each of which has 
its own racial peculiarities. These four areas are, (1) Eastern Norway, 
(2) Western Norway, (3) North-central Norway, (4) the Far North. 33 

Eastern Norway consists of the seven following provinces: Hedmark, 
Akershus, Ostfold, Vestfold, Opland, Buskerud, and Oslo. This section of 
the country, which includes the Oslofjord region and the long valleys 
which run towards the Dovre Mountains, forms a definite ethnic unit, 
within which much internal movement takes place between valleys and 
provinces, and much migration from the country districts to the city of 
Oslo. It has, however, but little to do with other parts of Norway, which 
are isolated from it by a number of barriers. The population of this eastern 
section is relatively uniform, both locally and as a whole. 

Although there seems to be an almost completely submerged brachy- 
cephalic element along the coast, this is not very much in evidence, for 
the main racial type of eastern Norway is a regular Halstatt Iron Age 

80 Bryn, H., and Schreiner, K. E., Somatologie der Norweger. 

81 Bryn, H., Homo Caesius; Bryn, H., AAnz, vol. 9, 1932, #2, pp. 141-164, and earlier 

Schreiner, Alette, Die Nord-Norweger; Anthropologische Lokaluntersuchungen in Norge; 
Voile, Hdlandsdaly und Eidfjord. 

32 For bibliography of these authors, see Bryn and Schreiner, pp. 607-608. 

33 Unless otherwise designated, the following pages are based upon Bryn and 


Nordic. This type, although predominant throughout the region, seems 
to be especially concentrated in the five valleys of 0sterdal, Gudbrandsdal, 
Valders, Hallingdal, and Numendal, forming parts of the three provinces 
of Hedmark, Opland, and Buskerud. Here, in a region almost unoccu- 
pied before the Iron Age, Bryn 34 believes to have found a refuge area of 
the classic Nordic race, with less admixture of other stocks than is the case 
elsewhere in Norway, or for that matter, in Europe. Hence his specifica- 
tions, both metrical and morphological, may serve as a standard of future 
comparison for use in the study of less typically Nordic populations. 

Army recruits from this region serve as a basis of study, while a series 
of farmers of old, indigenous ancestry forms a check series which repre- 
sents the original Iron Age population with a minimum of more recent 
admixture. These people must be considered tall, since the men attain 
in adult life the mean height of 172 cm., but from the Norwegian stand- 
point this stature is not unusual. In bodily proportions this type is rela- 
tively long legged and short bodied, moderately broad shouldered, and 
relatively short armed. The bones are typically small and fine, and the 
general musculature tends to leanness, while corpulence is very rare. On 
the whole the impression is given that the muscles lie close under the skin, 
and stand out in clear relief. A predominantly leptosome constitutional 
type seems to be characteristic. 

The mean vault dimensions of the recruits from these valleys are: 
length, 195 mm., breadth, 149 mm., and auricular height, 126 mm., with 
a cephalic index of 76.8. The native farmers are even longer headed, with 
a mean index of 75.5. Since these indices reflect figures of 73-75 on the 
skull, it may be readily seen that the original Iron Age Nordic vault form 
has been transferred to eastern Norway with little or no modification. 
Frontal and bigonial diameters average 105 mm., while the bizygomatic 
mean is 135 mm. The face height, given by Bryn as 122 mm., 35 is only 
moderately long. The nasal dimensions, of 55 or 56 mm. by 33.8, produce 
an index of 60 or 6 1. 36 

Ash-blond hair is typical of one-half of the native farmers, the rest 
having light brown and brown shades; only four per cent have hair that 
is black or dark brown. The rufous tinges of hair color are especially rare. 
Among the recruits, unselected as to provenience of ancestry, dark hair 
is twice as common, and the ash-blond shades are found in only one-third 
of the group. Thus we may, from this material, specify that the typical 

84 Bryn, H., AAnz, 1932; also Homo Caesius. 

86 It seems likely that Bryn located nasion a little lower than did A. Schreiner, judging 
by comparisons elsewhere. It is also likely that this mean should be nearer 125 or 126 

36 Here again I feel that Bryn's mean nose height of 54.7 mm. is a little too low, and 
that his nasal index of 61.8 is somewhat high. 


hair color of the living examples of the Iron Age Nordic race ranges from 
a medium brown to an ash-blond, with a minimum of rufosity, and a 
small brunet minority. 

The hair form most prevalent in Norway is a low waviness. Although 
low waves are characteristic of the southeastern valley country as well as 
of other regions, there is, nevertheless, a higher ratio of straight hair among 
this long-headed population than in other parts of Norway. Although the 
ratio is only 30 per cent, as against 66 per cent for the low- waved variety, 
yet these figures are so at variance with those for the rest of the kingdom 
that one may specify the hair form of this type as low-waved to straight. 
The beard is of but moderate abundance, although it increases consid- 
erably with age, and the body not especially hairy. 

In eye color as in hair color, the native farmers are lighter than the 
recruits, with 86.5 per cent of light and light-mixed eyes (Martin #12-16) 
as against 76 per cent. Of the recruits, 38.5 per cent have pure light eyes 
(Martin #15-16). This is by no means the lightest-eyed region in Norway. 
This material shows us what had been previously suspected, that the 
Nordic eye must be considered light mixed in typical form, rather than 
pure light. According to Bryn the commonest form of unpigmented eye 
found in this region is a light blue one, with large meshes and iris fibers 
set quite far apart, so that the iris pattern appears open. 

The skin color in this area, as in most of Norway, is almost invariably a 
pinkish white (von Luschan #3). This skin is of a fine texture; according 
to Bryn it is soft and easily punctured by a hypodermic needle. Owing 
to this thinness and the delicate quality of the skin, the cartilaginous and 
osseous structure of the face is often clearly discernible beneath it. 

The forehead of this type is for the most part sloping, forming a profile 
line parallel to that of the nose. It is medium to narrow in breadth, and, 
in comparison to other Norwegian types, relatively flat in both planes. 
The browridges are usually present, but are weakly developed, and the 
depression of the nasal root moderate. The nose may be described as thin, 
steep-walled, and high-bridged. In profile, it is for the most part straight 
or slightly convex, with a high incidence of wavy forms, and there is 
usually a noticeable transition between the bony and cartilaginous por- 
tions. Owing to the thinness of the skin, the line of suture between the 
two nasal bones may frequently be observed. The tip of the nose is thin, 
and for the most part raised slightly above the horizontal plane. The 
nasal wings are compressed, and the nostrils form long ovals, set at a very 
acute angle from one another. These nostrils are visible from the side, 
and slightly visible from in front. 

The bony orbit of the eye is rather high, and the eye normally quite 
wide open, with the upper lid reaching down over the upper quadrant of 


the iris, and the lower lid touching its rim. The eye slits themselves are 
horizontal, and are often partially covered, especially in old age, by a 
fold which hangs from the outer corner of the upper orbit. 

The eyebrows are thin, somewhat bowed, and seldom concurrent over 
the nasal bridge. The malars, small in size, are typically flattened in front. 
The zygomatic arches, however, are often bowed outward enough to give 
the face a pentagonoid appearance. This appearance is due to the flat- 
ness of the temples and the thinness of the soft parts of the arches, rather 
than to their skeletal prominence. 

The cheeks are in most cases thin, and the lower jaw long and deep, 
curving in front to a well-developed chin, with the gonial angles com- 
pressed and usually not visible. One of the outstanding features of this 
type, and of the Nordic race as a whole, is the great distance between 
the borders of the lower teeth and the point of the chin. The total im- 
pression of the face is that of a long, narrowish oval, often slightly rhom- 
boid, with prominent bony portions when seen in profile. The lips are 
usually thin, the mouth rather small, and the nasal sills well developed. 

The cranium itself is a long oval when seen from above, with almost 
parallel sides, and a marked transition from the frontal to the temporal 
bones. The greatest breadth is located as often in front of the center as 
behind it. Seen from the front, the cranium looks steep or parallel sided, 
and arched or vaulted on top. From the side, the contour of the head 
sweeps flatly back from a somewhat retreating forehead to a curved or 
projecting occiput. The highest point of the head is over the ears, and 
there is no pronounced tendency for either the forward or rear portion of 
the head to be higher than the other. Judging by gross bulk measure- 
ments the heads of individuals of this type may not be classed as large, 
nor high; their principal character is narrowness, a feature which con- 
tinues down to the face, and also to the nose. 

Although this distinctive type is today most concentrated in the long 
valleys of southeastern Norway, it is by no means confined to that region. 
It is found all over Norway in greater or lesser solution, as is to be ex- 
pected, since it is the racial type of the invaders who brought Iron Age 
civilization to Scandinavia. Besides this clearly differentiated Nordic 
type, there seem, however, to be various submerged minority elements in 
the eastern Norwegian population which are not limited to any one dis- 
trict, but are diffuse throughout. One is a shorter, somewhat darker and 
less dolichocephalic element which may in part represent an aboriginal 
coastal population, but which may, to a greater extent, consist rather of 
racial elements brought from central Europe in solution by the Iron Age 
Nordic invaders. Some of it, again, is undoubtedly descended from the 
thrall population brought from many parts of western Europe by the 


Vikings. The fact that it is shorter, darker, and less dolichocephalic than 
the more clearly designated Nordic type does not mean that it is very 
short, very round headed, or very dark. 

Besides this submerged element, or medley of elements, which is ex- 
tremely difficult to isolate, there is a third type, characterized especially 
by a broad face and a broad mandible, which may be attributed without 
question to recent Finnish influence. Finns settled here in the Grue 
district of Hedmark some 300 years ago, and have since been largely as- 
similated to Norwegian nationality and absorbed into the Norwegian 
population. Very few members of this colony still speak Finnish, or 
identify themselves as Finns. 

On the whole, despite these influences, the eastern provinces of Norway 
form, apart perhaps from Sweden, the most characteristic concentration 
area of the central Nordic racial form in the world. This residual enclave 
is directly descended from the Iron Age Nordic population which once 
occupied an immense area on the plains of central and eastern Europe 
and western Siberia, and which elsewhere has been replaced, altered, or 

Western Norway, the next section under consideration, includes the 
provinces of Telemark, Aust-Agder,. Vest-Agder, Rogaland, Hordaland, 
Sogn og-Fjordane, Bergen, and M0re. Within these provinces there are, 
in contrast to those farther east, considerable local differences; as a rule, 
many round-headed peoples live along the coast, while rnesocephals pre- 
dominate in the inland valleys. 

In the province of Rogaland the brachycephalic element reaches its 
maximum and here, in fact, is located its center of greatest concentration 
in all Norway. The inner nucleus of this brachycephalic area is Jaeren, 37 
a flat coastal plain, locally uniform in race, but regionally distinct. Here 
alone, in all of Norway, occur natural deposits of flint, and for this reason 
Jaeren must have been an important source of implement material for 
both Mesolithic and Neolithic peoples. On the plain the pre-Iron Age 
population must have been particularly dense. 

In Jaeren, Arbo found 82 per cent of brachycephals, 88 a ratio as high 
as that usual in southern Germany, and a mean cephalic index of 83.2. 
The three other districts of Rogaland, by comparison, have mean indices 
of 81-82. The Jaeren people form, as a whole, a very definite and easily 
observed type which has been most fully described by Larsen. 39 This type 
is most concentrated in the parishes of Haaland, H^iland, Klepp, and 

37 Bryn and Schreiner, pp. 431-449. Arbo's and other previous studies are covered in 
this section. 

88 Arbo, using the Broca system of parti tionment, included all indices of 80,1 and 
over, pooling Broca's sub-brachycephalic and brachycephalic classes. 

89 Larsen, C. F., Om Jaedertypen. 


Time. It has a large cranial vault of medium height, very broad, and of 
considerable length. Individual cephalic indices go as high as 90 or more, 
but the mode for the type as a whole is 84. The occiput, nearly vertical, 
often shows a slight degree of flattening. The temporal bones are weakly 
curved, but the parietal tuberosities are strong. The forehead is broad, 
only slightly curved, and quite high, and usually of but little slope. The 
browridges are, on the whole, of moderate size. The head exhibits from 
above a roundish, oval form; it is not an evolved planoccipital skull, al- 
though individual crania have a tendency in this direction. The face is 
notable for its breadth, both between the zygomata and in the mandible, 
which is frequently heavy and deep. The nasal profile is usually straight, 
but in one case out of six it is concave. The chin is pronounced, and some- 
times pointed. Together, the face and head give an impression of square- 
ness, owing to the prominence of frontal and parietal tuberosities, and to 
the breadth of the face and jaw. 

In pigmentation, these brachycephals are slightly less fair than the 
few dolichocephals found in the same region, but they are still predom- 
inantly blond. Eighty-one per cent have blue eyes, and only 3 per cent 
brown. Most of the hair is either light or medium brown; only 30 per 
cent have dark brown hair, and less than 2 per cent black. 

Correlations within the Rogaland population prove little. The few 
dolichocephals jare very little taller than the brachycephals, who are as 
tall as the eastern Norwegian Nordics, with a mean of 172 cm. Red hair 
and brown hair are associated with the highest cephalic index level, and 
the round-heads tend to have longer and heavier bodies, and broader and 
heavier faces, than the long heads. That the brachycephalic type in 
Jaeren is basically of light- mixed pigmentation is made especially clear 
by the fact that what few brunets there are in Rogaland run taller, longer- 
headed, and finer-nosed than the population as a whole. The Jaeren 
brachycephals, therefore, are not short and dark as often stated, 40 but arc 
tall and predominantly light-mixed people with large heads. There is 
no question here of a short, dark, brachycephalic population having been 
absorbed into a Nordic body, since the brachycephalic group in Jaeren 
is numerically the principal one. 

In Hordaland, north of Rogaland, one finds a continuation of the same 
contrast between coast and inland valleys which occurs farther south. 
The brachycephaly of Jaeren, which extends southwards into the two 
Agders, also stretches northward in an attenuated form into Mid-Horda- 
land, where it is gradually submerged in the mesocephalic population. 

40 Arbo, to whom this statement has often been attributed, stated merely that the 
Jaeren brachycephals were shorter and more brunet than their long-headed brethren. 
The differences are actually very slight. 


Secondary nuclei of brachycephaly occur sporadically farther north, 
notably in Sunnfjord on the northern bank of the great Sogn fjord, and in 
the coastal districts and islands of M0re. 

So great has been the interest in the coastal brachycephals of western 
Norway that the equal importance of the mesocephalic population living 
more typically in the inland valleys and mountains of this part of the 
country has been somewhat obscured. On the basis of the cephalic index 
alone, it would be easy to dismiss them as a transitional form between the 
Iron Age Nordics of the east and the Borreby type brachycephals of the 
coast, but a number of considerations make this disposal impossible. The 
western Norwegian mesocephals are taller, blonder, and larger headed 
than either of the two types mentioned. In these and in other respects, 
they form a special population of their own. 

In many districts of these provinces mean recruit statures of 175 cm. 
have been recorded, with a record mean of 178 cm. in the Voss district 
of Hordaland. The cephalic index of 78 or 79, which is so constant here, 
is not a composite of dolichocephals and brachycephals, but represents a 
truly mesocephalic condition. 

Mme. Schreincr, in order to study this special group in greater detail 
than the recruit material permits, selected the high mountain district of 
Valle in Setesdal, in the northern part of Aust-Agder; and also two iso- 
lated districts of Hordaland, Halandsdal and Eidfjord. 41 Of these three, 
Valle yielded the largest series, and the most extreme local form of the 
population under consideration. This site was especially chosen because 
it is probably the most secluded, most conservative place in all Norway; 
its inhabitants are still living in many respects in the saga period, and 
mingle little with outsiders. 

Valle was first settled in the second and third centuries of the present 
era, while a second wave of colonists arrived in the ninth. Since the dia- 
lect spoken in Valle is purely west Norwegian, we may assume that the 
present inhabitants represent a survival of that segment of the coastal 
population which, during the first millennium, forsook the shore for the 
mountains behind it. 

In body measurements the Valle people are large, although the mean 
stature of 174.7 cm. for one hundred adult males is not the greatest in this 
region. The women, with a mean of 160.0 cm., are much smaller. The 
sex difference in height, as in many other features, is particularly great 
here, and much greater than in Norway as a whole; it totals 14.1 cm. in 
Valle, as against 10.0 cm. in the entire country. The Valle people are, as a 
rule, heavy boned, and like the rest of the population of which they are a 
part, longer and heavier bodied than members of the eastern Nordic type. 

41 Schreiner, A , Valle ^ Halandsdal, und Eidfjard, 


The mean head length of the Valle males reaches the extreme figure of 
198 mm., considerably longer than that of the dolichocephalic eastern 
Norwegian Nordics; the breadth, 154.9 mm., is as great as that among 
many brachycephals, although in this case, in view of the exceptional 
head size, the resultant cephalic index mean is only 78.9. A mean head 
height of 125 mm. is, however, moderate. The face is large, with a mean 
nasion-menton height of 128.3 mm., and a bizygomatic breadth of 
142.9 mm. The forehead and jaw are broader, likewise, than in most of 
Norway, with means of 106.6 and 109.2 mm. 

By and large, the morphological observations bear out the impression 
of robusticity shown by the measurements; the forehead is often quite 
sloping, the browridges frequently heavy, the faces angular, the jaws firm 
and deep. In keeping with the cultural recessiveness of Valle, the palate 
and dental regions are large and primitive, in a mediaeval or Iron Age 
sense. The pigmentation is exclusively light or light mixed, for in Mme. 
Schreiner *s sample which included one-fourth of the total population, not 
a single brown eye nor head of black or dark brown hair was discovered. 
Among the men, 90 per cent of pure and nearly pure light eyes were found, 
with but 3 per cent dark mixed; among the women, as is frequently the 
case elsewhere, the light-eyed category is smaller than that of the men by a 
full 10 per cent. In hair color the Valle males show 40 per cent of ash- 
blond, an equal number of various shades of brown, and the remaining 
20 per cent of light golden blonds. 

Mme. Schreiner, as well as Arbo thirty years earlier, considered that the 
Valle people represent a retarded sample of the Viking population which 
lived in western Norway a thousand years ago, and this conclusion is based 
on geographical and ethnological as much as on racial grounds. If this be 
true, and there seems little reason to dispute it, then we may at last have 
found the living counterparts of the Iron Age crania which might, in many 
respects, have been those of Upper Palaeolithic men. Historically we 
know that from the Neolithic onward no racial types could have entered 
this region except for a pre-Iron Age Borreby-Megalithic-Corded blend, 
and, later, the Iron Age Nordic race which we have already seen in the 
provinces to the east. Both Arbo and Mme. Schreiner detected a minor 
element in the Valle population which was smaller and finer boned, and 
which was presumably Nordic in the Iron Age sense. 

The third section of Norway, usually designated as a racial center, is 
the north central group of three provinces, M0re, South Tr^ndelag, and 
North Tr^ndelag, with especial emphasis upon the two latter. The two 
Tr0ndelags include several great valleys: Namdal, Orkdal, Meldal, 
Galdal, and Tydal, and a number of large islands as well. To the south- 
east, this region is effectively blocked from contact with eastern Norway 


by the Dovre Mountains. During the twelfth and thirteenth centuries, 
this was the most populous and most important part of Norway, in which 
was located Nidaros, the capital of the Norse kings. This region was a 
center of Norwegian aristocracy, and a base for extensive Viking expedi- 
tions. As a result of these voyages, the whole Trondhjem region must 
have received a relatively large influx of foreign slaves and thralls, while 
in some of the valleys, Saxons and Bohemians were especially imported as 
skilled laborers. Tyrker, the famous Rhineland German who discovered 
the grapes on Vinland and made the New World's first wine, was probably 
one of these immigrants. 

The modern population of the Tr^ndelag region is notable in that it 
exceeds the rest of Norway in a number of important features. One is in 
stature, for the tallest provincial means are found here; another is in the 
height of the cranial vault, which reaches a mean of 128 mm.; a third is 
in the percentage of blue eyes, for this is the lightest-eyed region of Norway. 
The hair, by contrast, is by no means the blondest, but there are significant 
deficiencies of ash-blond, and excesses of golden and of brown. This type 
is also characterized by a considerable face length, with narrower bizy- 
gomatic and bigonial diameters than are found in Norway as a whole. 
The type which possesses the characters enumerated above is especially 
concentrated in South Trjzfndelag, and most strongly in the valley of 
Orkdal. The other districts of the two Tr0ndelags show a tendency for 
this special Nordic type to blend into the mesocephalic western form which 
reaches its culmination in Valle. 

Bryn has compared the Tr0ndelagen people in observations with the 
eastern Norwegian Nordics, and his contrast here is as valid in most re- 
spects for the western Norwegian mesocephals as for the Tr0ndelagen 
people themselves, since the latter two are morphologically much alike. 

The Trj^ndelagen population has the same proportion of dark hair as is 
found in the eastern Norwegian valleys; but differs from the classic Nordic 
type in a low ratio of ash-blond (26 per cent) and a correspondingly high 
proportion of golden and brown. The hair form in Tr^ndelagen is usually 
wavy; it is coarser, and more abundant on beard and body. 

Although the Trjzfndelagens are the two lightest-eyed provinces in 
Norway, their commonest iris type is very light mixed (Martin 13-14) 
rather than pure blue. According to Bryn, the typical Tr0nder iris is 
close grained and opaque, for the fibers are dense and closely imbricated. 
Bryn contrasts this iris type with that of eastern Norway. The skin, while 
as light in color as that of the eastern Nordics, is coarser in texture and 
much tougher. As a result of this density of the integument, the bony and 
cartilaginous parts of the face do not stand out in fine relief. 

The forehead of the special Tr^nder type is higher, broader, and much 


less sloping, and the profiles of the forehead and nose are not parallel, but 
form a distinct broken angle. Frontal bosses, which do not appear in the 
Eastern Valley type, are frequently found, and the temporal region is 
fuller. The transitions from frontal to temporal and frontal to parietal 
regions are smooth and difficult to find, whereas with the eastern type 
they are clearly marked. The nose of the Tr0nder type, while equally high 
or higher, is typically straight or convex, with many wavy or undulating 
profile forms. The side walls are less steep, and the transition from bone 
to cartilage difficult to find without palpation. The tip is somewhat 
thicker, especially in old age, and the wings less compressed. 

On the other hand, the zygomatic arches are less prominent than those 
of the eastern type. Not only are they somewhat more compressed, but, 
at the same time, the temporal region above them is broader and fuller, 
so that the lateral profile of the face falls usually in an unbroken sweep from 
the side of the head to the lower jaw line. As with the Eastern Valley type, 
the gonial angles are not noticeable. The cranium as a whole is shorter, 
higher, and more rounded, and the occiput less prominently curved. On 
the whole, the impression is given of a better filled, more rounded, and 
less angular head and face. If one leaves in the description of hair and 
integument, and adds a prominence of zygomata and of mandible, this 
description will apply to the other end type of the tall, mesocephalic 
population of western and central Norway, that of Valle. 

In reviewing the data on the coastal and mountain population of western 
and north central Norway, from Aust-Agder to North Trjzfndelagen, we 
find ample evidence of the major survival of a pre-Iron Age population. 
Within this population at least three elements are seen. 

(a) A tall, heavily built, large-headed type, with a stature of about 
170-172 cm.; the cephalic index is about 84, which would correspond to 
82 on the dry cranium; the face is broad, the jaw broad and heavy, the 
occiput often flattish, the skull square in appearance more frequently than 
round; the pigmentation is partly but not extremely blond, with light- 
mixed eyes, and hair which is medium brown to light brown on the golden 
side in the majority of cases. 

(b) An extremely tall, somewhat slenderer type, with a stature of 
174 crn.; mesocephalic, with a more moderate head size in length and 
breadth diameters, but with a vault attaining 128-130 mm, in auricular 
height, which is very great for living races; a long face, narrower in bi- 
zygomatic and bigonial widths than (a), and as narrow in these respects 
as that of Iron Age Nordics; heavier, with craggier facial features and 
thicker, coarser soft part anatomy than the Iron Age Nordics, in this re- 
spect approximating type (a); characterized in pigmentation by almost a 
totality of very light-mixed eyes, especially of the blue variety with a 


minimum of yellow and brown spotting; and by a brown or golden-brown 
to golden hair color range. 

(c) A type which in reference to (b) is equally tall, equally meso- 
cephalic, but lower-vaulted and larger in length and breadth dimensions 
of the vault; equally long-faced, but wide in both malar and gonial diam- 
eters, heavy-jawed, large faced; similar in pigment characters to (b), but 
not, in all regions, equally blond; large-bodied, rugged, and large-boned, 
with a great sex difference in stature. 

In all three of these, the later Iron Age Nordic element has blended. 
Despite this influence, type (a) in its concentrated form, as at Jaeren, 
seems to have reemerged as what appears to be a faithful replica of the 
Borreby race in its various forms, while (b), a blend of Corded with (a) 
and with elements of glacial age, forms a special and very characteristic 
and historically important Nordic sub-type. In both (a) and (b) the 
Borreby element, which entered Norway from Denmark during the Neo- 
lithic, is probably more important than the local post-glacial race of 
Palaeolithic tradition, remnants of which are probably masked in both, 
but appear in strongest solution in (c) . Individuals of type (c) may well 
recapitulate, in most essential features, Upper Palaeolithic Western 
European man. 

In any case, the deviation of the western and north-central Norwegians 
from the standard eastern Norwegian form is indicative of the absorption 
of the latter by pre-food-producing Scandinavian strains, as well as by 
pre-Iron Age Corded blood. The oft-stated and overemphasized resem- 
blance between the western Norwegians and central European Alpines 
reflects merely the common origin in the glacial period of Borreby and 
Alpine ancestors. The Alpines, however, have undergone modifications 
involving size reduction below the earlier form, while the Norwegian sur- 
vivors have retained their ancestral dimensions. 

For the purposes of classification, I, propose to lump the types (b) and 
(c) together, using Bryn's name of Tr^nder type, to designate all tall, 
coarsely built, mesocephalic blonds who show a predominance of Corded 
and Upper Palaeolithic elements, in contrast to the classic, finer Nordic 
type. This lumping may be justified by the supposition that (b) and (c) 
form but local end types of a larger population in which both are present 
but less distinct. 

The fourth Norwegian area which merits separate consideration is the 
Far North, including the provinces of Nordland, Troms, and Finnmark. 
In this region it is not the description and identification of a special type, 
but the interactions of several different ethnic elements, and their reactions 
to a rigorous environment, which are important. These elements are the 
Lapps, whom we have already discussed; the Kvaens, who are Finlanders 


of late arrival and who will be discussed in a later section of this chapter; 
and the Norwegians, most of whom are recent immigrants from other 
parts of the kingdom. 

At the beginning of the Norwegian historical period, HlUogaland, which 
included Nordland and the southern part of Troms up to Malangenfjord, 
was thickly settled with Norwegians who lived along the coast and es- 
pecially on the islands, and whose ancestors had come up in open boats in 
order to carry on fishing. These prehistoric settlers came from Tr^ndelag, 
M0re, and also from more southerly parts of the country. In the ninth 
century, at the latest, Norwegians from HMogaland sailed farther north 
to hunt walrus and to exchange goods with the Lapps. A number of them 
settled, and in the thirteenth century the whole coast of Finnmark had a 
scattered Norwegian population. In the sixteenth century, Finnmark con- 
tained at least 6000 Norwegians. 

For the next three hundred years the Bergen merchants held a trade 
monopoly which prevented private enterprise, and destroyed the incentive 
for northward migration. Many of the earlier settlers returned southward, 
leaving a shortage of workers. In consequence of this, King Christian V 
sent a mixed company of thieves, prostitutes, and other undesirables to 
the north country from southern Norway and from Denmark, in order to 
reenforce the population of the fishing villages. In 1815, however, Nor- 
wegians began coming north in large numbers, most of them from the 
southern part of the kingdom. Up until the eighteenth century, fishing 
and trade were almost the only occupations, but about that time agricul- 
ture was begun in the broad valleys of Nordland and Troms, and, under 
the influence of the newer settlers from the south, it became an important 
economic factor. 

During the fifty years which elapsed between 1869 and 1920, the popu- 
lation of the north country grew from fifty to ninety-eight thousand. That 
of Nordland increased 88 per cent, of Troms 92 per cent, and of Finnmark 
102 per cent. The bulk of this increase was caused by the influx of Nor- 
wegians. In 1920 Norwegians or people who considered themselves Nor- 
wegians constituted 99 per cent of the population in Nordland, 89 per cent 
in Troms, and 61 per cent in Finnmark. Since these figures include, 
especially in Troms and Finnmark, a number of mixtures between Norwe- 
gians and Lapps, Kvaens, and both, the Norwegian population of these 
provinces deviates from the means of the kingdom in several respects, 
especially in a lowering of stature and a heightening of the cephalic index. 
Kvaen influence may be detected most clearly in an excessive breadth of 
face and mandible. 

Norwegians of pure descent from immigrants born in the south-eastern 
provinces have retained their original stature and head form, as well as 


their high incidence of ash-blond hair, but they have been modified 
through an increased bigonial breadth and a decreased minimum frontal 
diameter. Mme. Schreiner, who has studied with great diligence a large 
series of North Norwegians of all ancestries, suggests that this condition 
may be a result of environmental influences which have caused a thicken- 
ing of the tympanic plate and the development of a palatal torus among 
most circumpolar peoples, including such varied groups as Eskimos, 
Lapps, and Icelanders. 42 

In studying the racial characters of the Norwegian people we have 
made use of a body of well-documented material, unique in Europe. By 
means of it we have been able to reconstruct a probable scheme of Nor- 
wegian racial history. There is one further source, however, which should 
not be overlooked, and that is the large corpus of Norse mythology and 
oral history. This source should not, as is commonly the case with folklore, 
be relegated to the ash-heap of what the scientist is wont to call mere litera- 
ture, since a careful study of the social attitudes, descriptions, and events 
so well recorded in the saga material shows that these documents agree 
with and supplement the findings of archaeology and of physical anthro- 
pology. Two sources which, in this regard are of especial value are the 
Rigsthula lay of the Poetic Eddaf* and the historical work of Snorre 
Sturlason, 44 a prominent political and scholastic figure in twelfth century 
Iceland. 45 

According to the Rigsthula, the social classes of the Norse people were 
begotten in a mythical and rather simple way. The early god Heimdal 
travelled about his domain in disguise, making use of the assumed name 
Rig. In this capacity he had sexual relations with three women, each of 
whom bore him children. The first woman gave birth to a brood of short, 
dark, and ugly offspring, who became thralls, and were relegated to 
agricultural toil and unskilled manual labor. The second produced the 
carls, large, healthy, red-faced, red-haired men, with big muscles, who 
became smiths and craftsmen, who performed skilled tasks, ,and who 
were also, in many cases, small land owners. The third woman was de- 
livered of the jarls, the aristocrats, tall, lean men with blond hair and hard, 
cold, snake-like eyes, who fought and practiced the use of weapons, hunted, 
played games, and did no work. 

The poet who described so vividly these three classes in the Norse popu- 
lation has given us a priceless picture of the people of Scandinavia during 

42 Hooton, E. A., AJPA, vol. 1, 1918, pp. 53-76. 
Schreiner, A., Die Nord-Norweger. 

Schreiner, K. E., ur Osteologie der Lappen^ vol. 1, pp. 161-177. 

43 Bellows, H. A., Poetic Edda (translation), pp. 201-216. 

44 Sturlason, Snorre, Heimskringla, edited by Erling Mousen, see esp. pp. 1-12. 
15 See also in this respect, Shetelig, Falk, and Gordon, Scandinavian Archaeology. 


the pre-Christian Iron Age, as he saw them. The thralls, landless serfs, 
were, in part, prisoners brought to Scandinavia by the Norse seafarers, but 
this explanation cannot apply to the thrall class as a whole. A three class 
system was an old Nordic institution, common to most Indo-European- 
speaking peoples, and it is unlikely that the Iron Age invaders from 
central Europe had entered Scandinavia without their henchmen. Part, 
at least, of the thrall class must be considered the descendants of Danubi- 
ans, Binaries, and Alpines who were imported by their more aristocratic 
overlords, and who formed, in solution with Nordic, the lower class of the 
original population. 

The carls find no ready counterpart in central Europe, and were prob- 
ably largely indigenous, the Bronze Age prototypes of the peoples of 
Jaeren, Tr^ndelag, and Valle. The physical attributes of these carls are 
clearly contrasted with the more purely Nordic description of the jarls, 
who formed obviously the upper class of the Iron Age invading group, 
including many of the bondi, or free land owners without title, and who 
were apparently a numerous body. 

Let us turn for a moment to consider the historical work of Snorre 
Sturlason. This erudite scholar deals with the gods as if they were men, 
and treats their mythical actions as history. His rationalization seems to 
have been uncannily accurate. In the first place, Asgard, the home of 
the gods, was a town on the northern shore of the Black Sea. These gods 
fought a people called the vanir, with whom they eventually agreed to 
exchange hostages. Odin, the king of the gods, agreed to take Frey and 
Freya, two of the vanir, and these were soon deified along with their hosts. 
The gods then left Asgard; and moved northwestward; they sojourned in 
Denmark, and passed without much ado into Sweden. This country be- 
came their main home, and Uppsala their chief center. Odin worship, 
which arose among their descendants, the kings and jarls, was centered 
especially in this neighborhood, and the worship of Frey and Freya as 
well. , 

Thor, who was a rough-and-tumble bucolic god, is little mentioned in 
this Asgardian history; he was apparently an earlier god and the especial 
deity of the coastal people of Norway. Odin was a sophisticated person- 
age, wearing a finely woven blue cape and carrying an iron spear; Thor, 
who clothed himself in skins, carried a hammer as his weapon, and drove 
about in a goat-drawn chariot. If we grant that Odin was the chief 
god brought in by the Iron Age invaders, and surrounded with their 
classically-inspired trappings of luxury, then Thor was apparently the 
god of the older people, of the carl class, and he represents in his per- 
son and attributes a blend between the robust Mesolithic hunters and 
fishermen, and the Megalithic and Corded people. His association 


with the last named is clearly shown by his devotion to the double- 
headed hammer, which was probably nothing more nor less than the 

The worshippers of Odin and Frey were especially interested in the 
horse; horse sacrifices were made to these gods, and to Frey was dedicated 
the cult of the embalmed horse's penis. In Norway the horse was replaced 
to a certain extent as a funeral object by the ship; and the ships were made 
by the carls, who had learned their craft from their Megalithic predeces- 
sors and ancestors. With the introduction of iron, ship-building flourished, 
and the Viking was nothing more nor less than a sea-going central Euro- 
pean Nordic, who had exchanged his horse for a steed suited to a new 
environment, with the cooperation of a vigorous body of indigenous 
craftsmen and warriors, into whose racial body his own group was soon 


Iceland 46 was first discovered by the Irish, but when this event took 
place we do not know. Our first reliable account of their voyages to Ice- 
land is the book of the Irish monk Dicuil, written in 825 A.D. At that time, 
and presumably for some years before, the only occupants of the island 
had been Irish hermits, who found their arctic retreat an excellent asylum 
from the ills of the world. It was probably from the Irish that the Norse- 
men obtained their knowledge of this island, before the motive had 
arrived for them to go there and live in it. 

Toward the end of the ninth century King Harald Fairhair united 
Norway under his own command, and then tried to extend his authority 
to the Norsemen living in the Orkneys and other outlying regions. As a 
result of his activities the noblemen who refused to submit sailed forth on 
Viking expeditions, and the Norse population in the British Isles increased. 
Iceland, however, being a country which was practically uninhabited, 
offered a ready refuge to these political malcontents, who comprised, it is 
said, the highest nobility of Norway. 

In 870 A.D. Ingolf Arnarsson first settled in Iceland, and a period of 
intensive colonization followed which lasted from 874 A.D. to 930 A.D. The 
high nobles, including kings, jarls, and peers of lesser rank, brought with 
them their entire households, consisting of wives, concubines, housecarls, 
and slaves. Four hundred such chiefs are mentioned in the Landnamabok, 
the unique document describing in detail the settlement of Iceland and the 
parti tionment of its land. Various estimates reckon the population at the 

46 The bulk of this section is derived from Harmesson, G., Korpermasse und Kbrperpro- 
portionen der Islander, and from Seltzer, C. C., The Physical Anthropology of the Mediaeval 
Icelanders, unpublished MS. in Peabody Museum. Author's permission. 


year 950 A.D. between the figures of 20,000 and 50,000. The lower figure 
is probably more nearly correct than the higher. At any rate, the chances 
are that the servants and other undistinguished persons made up the 
majority, and that although the proportion of noblemen was high, it was 
not high enough to predominate in a numerical sense. 

The Landnamabok names the homes of 1003 of these immigrants. Of 
them 846 came from Norway, 30 from Sweden, 1 from the Faroes, and 
126 from the British Isles. Of those coming directly from Norway, the 
homes of 461 are known, as follows: Nordland, 51; Tr^ndelag and M0re, 
95; Sogn og Fjordane, 128; Hordaland, 77; Rogaland 10 (3 from Jaeren); 
Agder, Telemark, Vestfold, 67; the eastern valleys, 33. Of 113 known 
homes in the British Isles, the list is: Ireland, 52, Scotland, 31, Hebrides, 
26, and Orkneys, 4. Thus the Norsemen who came from Norway came 
mostly from the coastal regions, and especially from Hordaland, Sogn og 
Fjordane and points northward. Few were from the eastern valley region 
and fewer from the brachycephalic nucleus in Rogaland. Those from the 
British Isles were presumably Norse who had not occupied their new 
homes long enough to lose their Norwegian identity. 

The Vikings who came from the British Isles brought with them Keltic- 
speaking slaves and concubines, who formed a considerable community 
and who are frequently mentioned in the sagas. Some of the leaders un- 
doubtedly had Irish mothers. The exact ratio of these people to the total 
population is, however, a matter of controversy. Hannesson, who has 
measured the living Icelanders, estimates the Irish and other Keltic ele- 
ments to have formed some 1 3 per cent of the whole. At any rate, since 
the tenth century no new immigrants have entered Iceland in any num- 
bers, and hence the living Icelanders are the direct and unassimilated 
descendants of the Viking settlers and of their retainers. 

In a total of 33 of the longer poems, 47 the bards who composed the sagas 
gave physical descriptions of 67 early Icelandic persons, all important 
and drawn mostly from the noble class. Of these 54 were called large or 
tall, and only 3 medium sized. In regard to hair quantity, 8 out of 9 men 
were said to have long hair, and one thick. Six out of seven men had 
curly hair, and one straight. The following hair colors were observed for 
19 males: gray 2, white 1, golden blond 2, blond 3, red 3, light brown 1, 
brown 4, black 3. One female was given black hair. Of three beard colors 
noticed, two were red and one gray. One man had blue eyes, and two 
women black. Although these observations do not form a statistically 
valid series or a random sample, yet they may be regarded as ample proof 
that the ancestors of the Icelanders were of variable pigmentation. Since 
the persons described were all of high rank, the chances are that most of 
Heinzel, R., SAWV, vol. 97, 1881, p. 107. 


them were pure Norwegians, and that the pigmentation map of western 
Norway was not very different a thousand years ago from what it is 

The modern Icelanders, with a mean stature of 173.6cm., are taller 
than most Norwegian groups, and come closest in general bulk to the 
Valle and Tr^ndelagen populations. In bodily proportions, too, they seem 
to be moderately thick-set and heavily muscled, and to be long spanned 
and relatively long bodied. In these general somatic characters they reveal 
the fact that their ancestors came more from the coast than from the in- 
terior of Norway. 

Their heads, being very long, with a mean of 197.3 mm., and rather 
broad (154.1 mm.), may be duplicated in size only in Valle, and in Ire- 
land. A head height of 126 mm. likewise fits into the general West Nor- 
wegian picture, as does a mean cephalic index of 78.1. 

The Icelanders, with a nasion-menton height of 130.1 mm., are very 
long faced, but their excess over the Norwegians in this character is partly 
a matter of technique. 48 They are actually not much longer in this char- 
acter than the people of Valle. The breadths of the face, the minimum 
frontal, bizygomatic, and bigonial (106.5, 140.6, and 108.5 mm.), are all 
broader than the corresponding dimensions in Norway as a whole, but 
they are comparable to those found in the provinces from which the Ice- 
landic ancestors came. The excess of the jaw breadth over that of the fore- 
head may indicate an adaptation resulting from rigorous dietary condi- 
tions, 49 as Mme. Schreiner also observed in northern Norway. 60 The 
noses are very high (58.8 mm.), and of moderate breadth, with a nasal 
index (60.2) on the lower border of leptorrhiny. One-half of the nasal 
profiles are straight, one-third concave; the remaining 17 per cent are 
mostly undulating, with a few convex. On the whole, less convexity is 
found here than in most districts of Norway or of Ireland. 

Hannesson, although he used the Fischer chart, divides his hair color 
categories in such a way that one cannot distinguish the ash-blond from the 
golden class. Other evidence, however, clearly indicates that, of the two, 
the latter is in the majority. Of pure blond hair (Fisher #12-24) he finds 
but .8 per cent as against 13.1 per cent for Norway, 61 and 5.5 per cent from 
Sogn og Fjordane, the province from which the largest number of settlers 
to Iceland came. In his light brown class (Fischer #7-11, 25-26), which 
includes what other authorities usually call ash-blond, he finds 52 per cent 

48 In recruit material used in the Somatologie nasion is quite apparently located lower 
than is consistent with either Hannesson's or Mme. Schreiner's techniques. A series of 
Icelanders measured by Ribbing includes a face height mean of 122 mm.; cf. Ribbing, 
L., LUA, N. F. Afd. 2, vol. 8, #6, 1912, pp. 1-8. 

49 Hooton, E. A., AJPA, loc. cit. 

60 Schreiner, A., Die Nord-Norweger. il Recalculated from Bryn and Schreiner. 


of his series, as compared to 64.8 per cent for Norway and 59.8 per cent for 
Sogn og Fjordane. 

Thus although the Icelanders are still prevailingly light haired, they are 
darker than any purely Norwegian population in Norway. In Norway 
black hair is everywhere, except among the Lapps, in a very small minor- 
ity; in Iceland it rises to the figure of 9 per cent, while red hair accounts 
for 3 per cent more. The presence of these two colors in such quantities is 
an excellent indication of the persistance of a strong Irish strain. 

This indication is strengthened by a study of Icelandic eye color. The 
ratio of light- and very light-mixed eyes (Martin #13-16) rises to 76 per 
cent, as high as the Tr^ndelagen ratios. But in Tr0ndelagen the majority 
are light-mixed eyes (Martin #13-14) while in Iceland, as in Ireland, 52 
over half of all eyes are pure blue. 

On the whole, the Icelanders represent a racial population which is 
most closely related to the mediaeval inhabitants of the western Norwegian 
coast, from Hordaland to Tr^ndelagen; they fit typologically into a mid- 
point between the two extremes of the Tr0nder category. They show 
certain developments of their own, particularly in their excessive face 
length, and in what seems to be an Arctic modification of the palate and 
jaws. In some respects they show perceptible Irish affinities; as in the re- 
tention of an excessive head size, and in the disharmony between very light 
eyes and hair of but intermediate blondness. In this series, even more than 
in the living Norwegian material, the resemblance to Upper Palaeolithic 
cranial and facial types is manifest. 53 

(6) SWEDEN 64 

Sweden, which occupies the more southerly, less mountainous, and 
larger side of the Scandinavian Peninsula, is in area the fifth largest coun- 
try in Europe. Most of its land is of high economic utility, since the low, 
well-watered slope of southern and central Sweden, dotted with lakes, is 
well suited for agriculture, while in the north, large forests and plentiful 
mineral deposits furnish materials for industry. Since 1775 Sweden's 
population has grown from two to six millions, not including the million 
and a half who have emigrated to the United States. Much of this increase 
has been fostered by the growth of industrial life, especially in the mining 
areas and in the cities. Central Sweden, in a belt reaching southwestward 

62 See Chapter X, section 2. 

63 Seltzer, G. C., op. cit. Seltzer finds a Cr6-Magnon-lifee type in a mediaeval cranial 
series from Haffiarderey, collected for the Peabody Museum by Vilhjalmur Stefansson. 
His opinion as to this resemblance is substantiated by both metrical and morphological 

64 The principal sources for this section are : 

Lundborg, H., and Linders, F. J., The Racial Characters of the Swedish Nation. 
Retzius, G., and Fiirst, G. M., Anthropologia Suecica. 


from Stockholm, and the peninsula of Sk&ne, are the regions of thickest 
settlement. Most of the Swedes who have gone to the United States 
originated in Gotaland, the southwestern part of the kingdom. 

In prehistoric times, Sweden, although less populous than Denmark, was 
far more important than Norway. From Ancylus times until the begin- 
ning of the Iron Age, the southwestern portion opposite the Danish Islands 
was a center of cultural activity, while the central and northern parts of 
the country were conservative and rustic cultural outposts. The brachy- 
cephalic Mesolithic population so typical of the Danish islands was less 
firmly rooted in Sweden, and the successive invasions of Megalithic and 
Corded people passed over into Sweden relatively unaltered, and produced 
a greater proportionate effect upon the racial composition of this country 
than upon that of Denmark. The Corded people, especially, moved north- 
ward into the central portions of the kingdom, and probably entered Tr0n- 
delagen, where their racial type is still important, by the Swedish route. 

The Iron Age invaders, the linguistic ancestors of the modern Scandi- 
navians, again chose Sweden as their especial sphere of colonization, and 
settled here in greater numbers than in Denmark or in Norway. Sweden 
became a great breeding ground for Nordic peoples, chief worshippers of 
Odin and of Frey, and after less than a thousand years, the country became 
so crowded with them that overpopulation, coupled with the onset of an 
adverse climate, forced a huge mass exodus southward. 

This movement was, in effect, the great series of Germanic migrations, 
the Volkerwanderung, which spread from Schleswig-Holstein and the 
Low Countries, on the west, and from the mouth of the Vistula on the 
east. The Goths, the Burgundians, and the Vandals, except for the Franks 
and Saxons, the most numerous and most important tribes of Germans, all 
had their origins in Sweden. As a womb of peoples Sweden was more 
important than Norway, and at an earlier date. Sweden was, in fact, to the 
continental world what Norway was to Britain, Iceland, and Normandy. 

Although, since the Iron Age, Sweden's historical role has been that of 
a feeder of peoples, she has at various times, and to a lesser extent, acted in 
the opposite capacity. During the Volkerwanderung the remnants of the 
Herulians and various bands of disappointed Goths returned to the Nordic 
homelands, tired of wandering, and it is not unlikely that they brought 
with them new racial elements picked up in Hungary and in the lands 
north of the Black Sea. Later on, during the Viking period of the ninth 
to eleventh centuries, Swedes, as well as Danes and Norwegians, raided 
many countries and brought back with them thralls from the British Isles, 
France, and the lands across the Baltic. According to Nordenstreng 65 

56 Nordenstreng, G., Origin, Growth, and Racial Components of the Swedish Nation^ in 
Lundborg and Linders, pp. 41-49. Special ref. to p. 44. 


these prisoners were settled most commonly in the present county of Upp- 
land, immediately north of the city of Stockholm. 

The development of cities in Sweden drew to that country large numbers 
of traders and merchants, from Viking times onward, and these commer- 
cial people were largely of Germanic origin. Frisian and Saxon chapmen 
were the first, and these were followed by others, in later times, from vari- 
ous parts of Germany, including the southern principalities. During the 
period of Sweden's great military expansion (1611-1718 A.D.), when the 
kingdom extended over large parts of Germany, many Germans were 
made noblemen, and went to live in Sweden. Thus the German blood in 
Sweden is a factor to be reckoned with, and has influenced, chiefly, 
the city population and the nobility. The latter class has also received 
strong infusions from Scotland, for Scotsmen, who served under Gustavus 
Adolphus in large numbers, were in many instances rewarded for their 
bravery by elevation to the Swedish peerage. Furthermore, Walloons, 
who represented a much darker and rounder- headed racial element than 
these other immigrants, were brought to Sweden during the seventeenth 
century to work in the iron foundries. Some thirty or forty thousand of 
their descendants can still be identified. 

More important than any of these absorptions, in all likelihood, has 
been the influence of the Finns upon the Swedish people. In the Middle 
Ages, Kvaens wandered into the northern counties, but not in great num- 
bers. The same Kvaenish migration which affected the northern provinces 
of Norway from 1700 A.D. onward, also reenforced this element in northern 
Sweden. During the sixteenth and seventeenth centuries, other Finns 
settled in Varmland and Dalarne, counties bordering on the Norwegian 
provinces of Ostfold and Hedmark, and the Finns of Grue 56 in Norway 
came as part of this same migration. Other Finns remained in scattered 
settlements between the Varmland and Dalarne nucleus and the head of 
the Gulf of Bothnia, while still others penetrated as far south as Stockholm. 

Although this migration ceased about 1700, over 13,000 Finns had come 
to Sweden and to a small district in Norway. Although these Finns were 
not numerous, the population of Sweden at that time was no more than 
one and a half millions, and the Finns were particularly prolific. Today 
only .two villages in Varmland retain Finnish speech from the time of this 
migration. In Norrbotten, in the valleys of Torne and Muonio, more 
recent colonies of Finns, from southwestern Finland, still speak their own 
language, and form a distinct alien bloc. In all there are, at present, 
about 30,000 Finnish speakers in Sweden, in addition to whom it is esti- 
mated that well over 100,000 Swedes are at least partially of Finnish 
66 See p. 313. 


In comparison with most European countries, Sweden has, in post-Iron 
Age times, been subjected to remarkably few foreign influences which 
would affect her racial composition. Despite the absorptions and immigra- 
tions noted above, Sweden remains one of the most homogeneous nations 
in Europe both in race and in pedigree. This homogeneity is largely the 
result of geography, for in contrast to the rugged Norwegian landscape, 
with its mountains and fjords and distinct centers of racial concentration, 
the flat surface of Sweden, with its modern industrial development and 
fluidity of population, has brought about a striking racial unity. In 
Sweden social and occupational differences in physical type are almost as 
great as regional ones. In no racial character are Swedish sub-groups, 
whether geographical or social, strongly differentiated. 

The same basic Hallstatt Nordic type which found such a favorable 
breeding ground in Sweden during the Iron Age is still the predominant 
race in that kingdom. It has absorbed into its ethnic body both older and 
newer peoples, and has spread the resultant blend with remarkable even- 
ness over the surface of the nation. On the whole, Sweden is the most 
Nordic nation in Europe in the Iron Age sense, and it is much more Nordic 
than Norway. At the same time, owing to geographical factors again, the 
valleys of southeastern Norway contain as unaltered an Iron Age Nordic 
population as any in Sweden. The metrical characters of the recruit 
material for the entire Swedish nation are very similar, in fact, to those of 
the southeastern Norwegians. 67 The stature mean of the Swedes is 
172.2 cm., and their characteristic bodily proportions are equally close to 
the Norwegian standard. Regional variation in stature stretches only 
from 169.9 cm. in the northeastern manufacturing districts to 172.5 cm. 
in the central provinces contiguous with Tr0ndelag. In the far north, 
where Finnish influence is common, and in the south, where the older, 
more brachycephalic populations of the Neolithic and Bronze Ages were 
seated, the length of the trunk is relatively greater, and of the legs smaller, 
than in the central parts of the kingdom, but these regional differences are 
less pronounced than those between social and occupational groups in the 
nation as a whole. As in Norway, the population drawn to the cities is 
notably shorter-armed than that which remains upon the land. 

The mean head length of Swedish recruits is 193.8 mm., and the 
breadth 152.3 mm., yielding a cephalic index of 77.7. The longest heads, 
with regional means running up to 195 mm., are found in the west, over 
against Norway, and the shortest in the north. The lowest cephalic index 
mean is 76.7, and the highest, concentrated in the north, are all below 80. 
The three principal breadth diameters of the face, minimum frontal, 
bizygomatic, and bigonial, have national means of 104.6 mm., 136.0 mm., 

67 Lundborg and Linders, op. cit. 


and 103.4 mm., respectively, all of which are typically Nordic and com- 
parable to those of the eastern valley Norwegians. Slight regional differ- 
ences place the narrowest foreheads and faces in the western counties, and 
the broadest in the north and south. The total face height of 126,6 mm. 
is again a typically Nordic mean, comparable to that obtained by Bryn 
in his later work on the Eastern Valley people. 58 While the narrowest faces 
are found in western Sweden, as is to be expected, the longest are typical 
of farmers in the north, where the Corded element may be slightly more 
prevalent. The Swedes are typically leptorrhine, 69 and the commonest 
nasal profile form is straight. Concave noses, which reach the rather high 
figure of 28 per cent in the kingdom, are commonest in the north and 
least frequent in the south. 

According to the Anthropologia Suecica, 52 per cent of Swedes had 
ash-blond hair, and 23 per cent golden. Thus the proportions of these 
two classes of blondism are reversed in comparison to Norway. The two 
countries are about equal in amount of dark hair shades, but, by and large, 
Norway would seem to be lighter haired than Sweden, 60 if we may rely 
upon a comparison based on a correlation of two scales. In any case, the 
most numerous category is a medium to light brown, with extreme blonds 
in the minority. Regional differences, though slight, are suggestive. 
Gotaland, the Goth country, as southern and southwestern Sweden was 
anciently designated, is lighter than Svealand, or central Sweden; Norr- 
land, the north country, is in turn the darkest. The most red hair is found 
in the west and south, and the least in the east, toward Finland. 

Retzius and Fiirst found 67 per cent of light eyes, 29 per cent of mixed, 

68 Bryn, H., AAnz, vol. 9, 1932, pp. 141-164. It is higher than the Norwegian recruit 
material means, which were apparently taken with a different technique. 

69 The only nasal constants in the L. and L. material are for Skaraborgs Ian, where a 
N. I. of 62.7 is found. The nasal dimensions of 61.37 mm. for height and 30.18 for 
breadth (p. 102) are presumably misprints. 

60 This statement is in direct contradiction to the opinion of most anthropologists, 
especially of W. Scheldt, as expressed in his Die rassischen Verhdltnuse in Nordeuropa, 
(ZFMA, vol. 28, 1930, pp. 1-198) and is by no means certain It is based on the fol- 
lowing correlation of the L. and L. material with that from the Somatologie der Norweger: 

12-25 flaxen 6.9% 27.9% 

7-11,26 light brown 62.5 50.0 

5-6 brown (medium) 25.1 17.2 

4 brownish black 2.0 3.7 

27-28 black .2 .1 

1-3 red 3.3 1.3 

The Swedish recruits were observed for hair color by means of a local chart, which was 
later correlated with the Fischer standard. (L. L., p. 10.) The c6mparison between 
the Swedish and Norwegian results was made by recombining the total Norwegian 
series according to the Swedish divisions. The difference in amounts of red is un- 
doubtedly due to a difference of standards, as Conitzer has previously stated. (Co- 
nitzer, H., ZFMA, vol. 19, 1931, pp. 83-147.) 


and 4 per cent of dark. In the first category were presumably included 
light eyes with a slight spotting, as in the Martin numbers 13 and 14. 
The Lundborg and Linders study, made with a different observational 
scheme, 61 raised the first category to 87 per cent, and the third to 5 per 
cent. In any case, there can be no doubt that the eye colors of the Swedish 
people are predominantly light mixed and light, as in Norway; and that 
the lightest eyes in the kingdom are found in western Sweden, and the 
darkest in the north. 

Correlations within the Lundborg and Linders series of 47,000 men show 
certain slight linkages, which could be dismissed as insignificant if found 
on smaller samples. The cephalic index decreases slightly, and the facial 
index rises, with an increase in stature; similarly, the tallest statures have a 
tendency to go with brown hair and light eyes. It is not unreasonable to 
suppose that this combination may be a faint reflection of the absorption 
of a Corded racial element into the population of Sweden. In the same 
way an association of flaxen hair, moderate stature, mesocephalic head 
form, and convexity of nasal profile, makes it unlikely that all high cephalic 
indices in Sweden are due to East Baltic influence, and suggests rather a 
survival of mesocephalic and brachycephalic elements in southern Sweden, 
comparable to those in western Norway. Truly short stature, linked with 
dark pigmentation and round head form, furnishes an infrequent com- 
bination, but one which may imply a Lappish strain in the far north, sub- 
merged Alpine elements, or both. 

The Swedish material, and especially the correlations, confirms the 
opinion formed in Norway, that the Nordic race as such is not and was 
never wholly blond. The characteristic eye color is blue or gray, and the 
presence or absence of a small amount of superficial iris pigment seems 
racially irrelevant. At the same time, it is likely that all hair color shades 
from a light medium brown to the lightest, whether on the ashen or golden 
side, should be considered as "pure" lights, since, as the Swedish material 
shows, persons having these shades on the head have, as a rule, the same 
colored pubic hair. In Sweden, as in Norway, what linkages there are 
which point to the survival or resegregation of a Corded type indicate 
that this type was characterized by exceptionally light eyes, but a pre- 
dominantly brown shade of hair. 

Abundant anthropometric data from Sweden make it clear that the 
basic, and by far the most numerous element in the population is, as in 
eastern Norway, an Iron Age Nordic one, transferred from its central and 
eastern European home; earlier elements have survived less here than in 

6i " #1 -s light iris (blue, gray, pale yellow, or green), also light iris with insignificant 
brown spots, points, or patches; 2 mixed iris and light iris with brown aureole; 
3 = light brown or dark iris." L. & L., p. 10. 


Norway. There is, however, a strong concentration of unreduced Briinn 
and Borreby types, as illustrated in plates 4 and 5, in the fishing and sea- 
faring population of the southwestern coast, across from Denmark; the 
presence of these types, although not clearly indicated by existing surveys, 
cannot, nevertheless, be denied. 

At the same time, Corded elements within the Nordic racial body are 
most evident in the north, and especially near the Norwegian provinces 
of Tr0ndelagen. Lappish influences are also to be felt in the far north, 
while modern Finnish invasions and infiltrations have introduced the 
East Baltic type into central Sweden in some numbers. The nature of this 
type need not be discussed here, but will be studied in later sections of 
the present chapter. 

(7) DENMARK 62 

Denmark, the smallest and most southerly of the three Scandinavian 
kingdoms, is also the most densely populated, being inhabited by two and 
one half millions of people. It consists of the peninsula of Jutland, the 
isthmus of Schleswig, acquired since the World War, and the Danish 
archipelago. These islands, the largest of which are Zealand, Funen, 
Laaland, Falster, Moen, Langeland, and Sams0, although smaller in total 
area than the mainland, contain the bulk of the population. The island 
of Bornholm, situated to the southeast of Skane, is likewise Danish terri- 
tory, as are the islands of Lesso and Anholt, which lie in the midst of the 
Cattegat. On the southwestern coast of Denmark the Frisian Islands begin 
their chain, which is only broken by the mouth of the Elbe in its stretch 
from Denmark to Holland. Some of these islands are Danish, some are 
German, and others are Dutch in nationality. Far separated from Den- 
mark, but under its sovereignty, lie the Faroe Islands, between the Shet- 
lands and Iceland, and Iceland itself is an autonomous state under the 
Danish crown, while Greenland, a restricted crown colony, is the home of 
a few thousand Danes. 

Throughout the prehistoric period Denmark was the cultural center of 

62 The principal sources for the physical anthropology of the living in Denmark are : 
Bardenfleth, K. S., MODA, vol. 3, 1929, pp. 3-49. 
Burrau, C., MODA, vol. 1, 1907-11, pp. 243-260, 277-284. 
Hannesson, G., op. cit. 

Hansen, Andreas M., NMN, vol. 53, pp. 202-266. 

Hansen, S0ren, MODA, vol. 1, 1907-11, pp. 69-81, 204-220, 222-240, 287-307; 
vol. 2, 1920-28, pp. 363-389. 

Hansen, Sjzfren, and Topinard, P., RDAP, vol. 3, 1888, pp. 39-41. 
Heiberg, P., MODA, vol. 2, 1920-28, pp. 296-300, 353-360. 
Mackenprang, E., MODA, vol. 1, 1907-11, pp. 11-68. 
Ribbing, L., MODA, vol. 1, 1907-11, pp. 193-202. 
Steensby, H. P., MODA, vol. 1, 1907-11, pp. 85-148. 
Westergaard, H., MODA, vol. 1, 1907-11, pp. 353-391. 


Scandinavia, and likewise the center of greatest population. The profusion 
of Neolithic and Bronze Age monuments and graves shows that before the 
Iron Age invasions both the mainland and the islands were densely in- 
habited; in view of this crowding, it is not surprising that the newcomers 
found greater room for expansion in Sweden and eastern Norway. From 
Ertebjzflle times onward the Danish Islands, and to a lesser extent the main- 
land, was the focal point in northern Europe for the settlement of the 
brachycephalic Borreby people. With them had mingled Megalithic sea- 
farers in large numbers, while the Corded people had concentrated their 
activities on the mainland. It is not surprising; therefore, that a population 
so firmly attached to its milieu as that of pre-Iron Age Denmark should 
have survived the vicissitudes of centuries and eventually have reemerged 
in considerable strength. That this is exactly what has happened is the 
sense of the present section. 

During the Iron Age Denmark continued in its cultural leadership of 
Scandinavia, owing largely to its greater proximity to the source of civilized 
influences farther south, for Denmark was greatly affected by the reper- 
cussions of Roman civilization. In the Volkerwanderung period, Den- 
mark, furthermore, contributed heavily to the stream of migration south- 
ward; the Cimbri, the first Germanic people to come under the eyes of 
Rome, were natives of Jutland; the Jutes and the Angles who settled 
England with the Saxons from Schleswig-Holstein again came from 
Denmark. The later inroads of Danes into Britain strengthened the earlier 
contingents. Hence, Denmark played an even greater part in the settle- 
ment of the British Isles than did Norway. 

In contrast to Norway and Sweden, existing documents which cover the 
physical anthropology of the living Danes are scattered and incomplete. 
It is not possible to study the distribution of characters from village to vil- 
lage and county to county, nor to examine the special racial attributes of 
individuals. It is possible, however, to make a few general observations, 
and to supplement these with deductions based on common knowledge. 
In the first place, the Danes are not as tall as the Swedes and Norwegians, 
although their king is the tallest monarch in Europe. The mean stature of 
twenty-one year old recruits in 1925 was 169.4 cm., which varied between 
172.3 cm., on the island of Anholt in the middle of the Cattegat, and 
167.1 cm. for Fan0, the northernmost of the Frisian Isles. In general, 
Jutland and Schleswig are comparatively tall, with mean statures of 
170 cm., while the island population is a centimeter or two shorter, es- 
pecially on Sams0, southern and eastern Zealand, Laaland, Falster, and 
Moen. Copenhagen and the adjoining counties of northern Zealand are, 
by contrast, quite tall. 

Aside from stature, there is no metric character in which all of Denmark 


has been regionally studied. In other measurements and indices one is 
obliged to refer to material which covers the country as a unit, or certain 
sections of it only. Data referring to bodily build indicate that the Danes 
are longer armed, wider spanned, longer trunked, and, in general, more 
heavily built than the common run of other Scandinavians, and resemble 
in these respects the western Norwegians more than any other group. 
Several series show that the mean head lengths of Danes in various parts 
of the kingdom are uniformly 194 mm., as long as the Swedish national 
mean, and comparable to that of the mesocephalic population of western 
Norway; variations in cephalic index are dependent rather upon variations 
in head breadth, which ranges from 154.7 mm. on the island of Bornholm 
to 158.8 mm. in the northern part of Sams0. That the higher cephalic 
indices in Denmark result from greater breadths instead of from lesser 
lengths, is a sure indication that we are dealing with a Borreby form of 

The mean cephalic index of Denmark, however, is but 80. 6 ; 63 and this 
sub-brachycephalic mean condition is not subject to much regional varia- 
tion. Although Denmark is the least long headed of the three Scandinavian 
kingdoms, nowhere in it may be found a regional population as round 
headed as that of Jaeren. Denmark, like Sweden, is flat and lacks natural 
barriers; one must expect a great national uniformity. The highest means 
yet recorded are 81.8 for northern Sams0, 81.4 for western Jutland, and 
for the isle of Anholt. No regional mean is under 80. 

Facial measurements on Danes are extremely rare; what there are show 
breadth diameters high for Scandinavia. Hannesson, in a small series of 
Danish sailors, finds a minimum frontal of 106.5 mm., a bizygomatic of 
139.5 mm., and a bigonial of 107 mm. In northern Sams0, an unusually 
brachycephalic area, the bizygomatic rises to 142.5 mm. Thus the Danish 
facial breadths resemble those found in coastal Norway, especially the 
rounder- headed districts, and in Iceland. 

Data on the hair and eye color of Danes is as extensive as that on 
stature, and covers the entire kingdom. Although no scales were used, 
the categories employed seem clearly defined and there can be little doubt 
as to the character of Danish pigmentation. Hansen found that "fair" 
hair decreased from 52 per cent at the age of 6 years to 33 per cent at 14, 
and fell to 16.6 per cent at the recruit age of 20 years. This "fair" category 
must, therefore, include pronounced degrees of blondism only, and ex- 
clude the light brown hues often designated as blond elsewhere. On the 
island of Sams0 Bardenfleth found only 7.5 per cent of hair which he was 
willing to call light, and 40 per cent of medium, 43 per cent of dark, and 
9 per cent of black. Sams0 is one of the darkest-haired regions of Denmark. 
68 Hansen has 80.6, Burrau 80.69. 


Judging from the distribution of the school children material, the southern 
part of the Danish mainland, toward Schleswig-Holstein, is the blondest 
section of the country; two regions are darkest: Thirsted, the northwestern 
county of Jutland, and the islands. 

What appears to be the most accurate division of eye colors is that of 
Bardenfleth, who finds 38 per cent of light, 59 per cent of mixed, and 3 per 
cent of dark eyes on Sams0. This is comparable to the eye color situation 
elsewhere in Scandinavia. Sams0 is one of the darker-eyed sections of 
Denmark, and regional eye color variations, though not great, follow those 
of hair color. 

In its available form, the Danish material is not so arranged that many 
correlations and regressions can be made from it. In Sams0, light-haired 
individuals are a half centimeter taller than dark-haired ones, and slightly 
higher in cephalic index. This regression runs counter to the slight geo- 
graphical association between darker hair, shorter stature, and rounder 
heads, from which racial inferences have been deduced. The associations 
noted in Sams0, however, agree with the similar correlations found in 
southern Sweden, which would point to the presence on both sides of the 
Cattegat of a special tall, blond brachycephal, particularly common among 
Swedish immigrants to the United States where the vulgar term "square- 
head 55 is used to designate it. Popular, subjective labels in the designation 
of races, used among persons ignorant of the existence of physical anthro- 
pology, are often truer than the hesitant results of erudite wanderings in 
the labyrinth of numbers. 

A knowledge of the racial history of Denmark, and a familiarity with the 
appearance of modern Danes, makes the interpretation of existing data, 
however fragmentary, possible. On the whole, the Danes form, as Burrau 
feels, a composite type which is inextricably blended, but which shows in 
individual variations leanings toward different ancestral forms, as well as 
toward new combinations. The blond /'square-head 5 ' noted above is an 
important type, heavy-boned and sturdy, basically Borreby in inspiration. 

The minority of brunet pigmentation, in Denmark not associated with 
brachycephaly, reminds one that the Danish Islands held the greatest 
concentration of Megalithic people in the whole north, and that these 
Megalithic people blended with the Borreby aborigines before the arrival 
of either Corded folk or Iron Age Nordics. On the whole, Denmark, like 
Sweden and Norway, may be called a Nordic country, but Nordic only in 
the modern Scandinavian sense. 

Before leaving the description of the living Danish people, two special 
problems remain, the racial character of the island of Bornholm, to the 
east of Denmark proper in the Baltic, and that of the Faroes. Ribbing, in 
a study of the Bornholm people, finds them taller, fairer, and somewhat 


longer headed than most of the Danes, and considers that they are most 
closely related to the southern Swedes inhabiting the island of Gotland. 

The Faroes, isolated in the northern seas between the Shetlands and 
Iceland, preserve a picturesque and mediaeval Danish population of 
fishermen. 64 These islands were first inhabited by the Scotch, who may or 
may not have left before the coming of the Vikings, which took place 
shortly before the settlement of Iceland. The Faroe males are as tall as 
Danes (169-170 cm.), and about the same in head form. (C. I. 79. 6.) 65 
The faces are distinguished by a considerable breadth of the mandible, 66 
found also in Iceland and among the northernmost Norwegians. Until more 
extensive information appears than that at present available, we may con- 
sider the Faroe Islanders typical descendants of Viking Age Danes and 
coastal Norwegians. 

In all three Scandinavian kingdoms, changes have been observed in 
stature during the last century. The normal amount of increase in young 
men draughted for recruiting has been somewhere between 6 and 8 cm. 
It would appear that one hundred years ago Danes of military age were 
only 164 cm. tall, on the average, while Swedes and Norwegians varied 
regionally between 166 and 168 cm. If one recalls the statures of the 
inhabitants of these countries before and during the pre-Christian Iron 
Age, it will at once appear that this increase has been actually a process 
of returning, under new stimuli, to an older condition. The depletion of 
these countries during the Volkerwanderung and the adverse climatic 
conditions of the Middle Ages must have had in the first instance a selective, 
in the second a depressing, effect upon national stature. 

In all three countries comparisons between city and country populations 
show that there is a tendency for the Iron Age Nordic type to be drawn to 
the cities, and to be, in general, the most restless element in the population; 
undoubtedly because it was the last to arrive and because it formed in 
many regions the upper social stratum. For these reasons again it is not 

64 Chief works on the Faroes are : 

Annandale, N., TRSE, vol. 25, 1906, pp. 2-24. Arbo, C. O. E., DGT, vol. 12, 
1893-94, pp. 7-14. 

Hansen, S0ren, JRAI, vol. 42, 1912, pp. 485-492; DGT, vol. 21, 1912, pp. 251- 
256; vol. 25, 1920, pp. 53-54. 

66 Considerable confusion is extant concerning the head form and stature of the 
Faroe Islanders. Arbo (1893) measured a series of 20 men from the northernmost and 
20 others from the southernmost island. He found that the stature and C. I. of the first 
group were 169.5 cm. and 75.2; of the second, 165.2 cm. and 83.2. His series of 60 men 
from Thorshavn fell into an intermediate position, approximating the means above 
given. These latter are taken from Hansen's series of 493 males from Suder0, and from 
Arbo's Thorshavn series. The startling regional differences of Arbo's work may be 
attributed partly to the small size of his samples, partly to the chance selection of iso- 
lated family groups. 

M Annandale's mean bigonial diameter on 20 men is 111.8 mm. 


inconceivable that the Volkerwanderung drained off this element in dis- 
proportionate numbers, and that the reemergence of older forms has been 
a result of this process, especially in Denmark, in western Norway, and in 
southern Sweden, where the older forms were originally most numerous. 
The three Scandinavian kingdoms, and especially eastern Norway and 
Sweden as a whole, remain the greatest single reservoir of the Iron Age 
Nordic race, but it is conceivable that that race was numerically more 
important in Scandinavia at the time of Christ than it is today. 


The next step in our survey of the living peoples of northern Europe 
leads us from Scandinavia, the present Nordic homeland, across the Baltic 
Sea to the countries in which the East Baltic race is most characteristic; 
the four republics of Finland, Esthonia, Latvia, and Lithuania. We pro- 
pose to study first the two northernmost, in which languages of the Finno- 
Ugrian family are spoken. But this will be done in a roundabout fashion, 
since before the racial history of the Baltic Finns may be fully understood, 
it will be necessary to deal with the entire ethnic and linguistic group of 
which they form a part. For this reason it will also be necessary to inter- 
rupt the geographical order tentatively followed, and to start with the 
Finnish homelands in eastern Russia. In an earlier chapter (Chapter VII, 
section 1), a survey was made of the skeletal remains of early Finno- 
Ugrian-speaking peoples of this region; some mention was also made con- 
cerning their early ethnic movements. It is the purpose of the present 
section to explain in a little more detail the linguistic and historic relation- 
ships of modern Finno-Ugrian-speaking peoples. 67 

For the sake of clarity, we will repeat that Finno-Ugrian, along with 
Samoyedic, forms the Uralic linguistic stock or sub-stock which may or 
may not be united with Turkic, Tungusic, and Mongolian in a Ural- 
Altaic supcrstock. It is now believed that early Finno-Ugrian was one of 
the two elements which blended to form basic Indo-European. 68 Although 
today their use is not as extensive as that of Indo-European, modern 
Finno-Ugrian languages are by no means archaic, and show no tendency 
toward disappearance. They are, in fact, spoken by a large number of 
peoples, living under extremely variable environmental and cultural con- 
ditions, and scattered over a wide expanse of territory. In three nations, 

67 Useful sources are : 

Atlas of Finland, 1925, esp. Wichman, Y., pp. 19-22. 
Jochelson, W., The Peoples of Asiatic Russia, esp. pp. 1621. 
Sirelius, U. T., The Genealogy of the Finns. 

Zolotarev, D. A., Etnickeskil Sostav Nasalenifa Sev.^ap. Oblasti i KareVskot ASSR; 
ibid., TKIP, vol. 15, #2, 1928, pp. 1-26. 

68 See Chapter VI, sections 1 and 8, Chapter VII, sections 1 and 3. 



Finland, Esthonia, and Hungary, divisions of Finno-Ugrian are the 
official languages, spoken by millions of people. In western Siberia, as 
well as in these countries, Finno-Ugrian speech occupies a large space on 
the map, but in this wilderness of forest and swamp it is actually spoken 
by very few persons. Elsewhere, throughout eastern central Russia and 

MAP 10 

This docs not include Osmanli Turkish as spoken in the former Turkish Empire. 
All Turkish speech is represented by crescents, Mongols by cross-hatching, and Sam- 
oyedic by small circles. Finno-Ugrian is represented by various types of lines and 
stipples, except for Lappish, which is indicated by crosses, and Livonian, which is 
solid. The northern instances of Carelian are Kvaenish. 

thence in a narrow band across to the mouth of the Gulf of Finland, it is 
found in little islands standing out in the midst of Slavic territory. 

Linguistic affiliations within the stock may best be illustrated by Ka- 
java's chart, reproduced in slightly altered form below. 69 In this chart the 
term Ingrians is used to include the various groups of Finnic speakers native 
to the Leningrad region; namely, the Vodes, the Lyds or Ijores, and the 

Kajava, Y., EA, #8, #9, 1922, pp. 353-358. 


two tribes called Evremeiset and Savakot, who live among Russians in the 
city of Leningrad itself. 70 

The early home of the united Finno-Ugrians is supposed by linguists to 
have been in the region which extends from the headwaters of the Dnieper 
and the western Dvina to the western slope of the Ural Mountains. The 
country around the Oka, the bend of the Volga, and the Kama are thought 
to have been occupied by Finno-Ugrians by the time that some of their 
southeastern tribes mingled with Caucasic-speaking peoples to produce 
I ndo-European . 

In their early home, during the first millennium B.C., 71 the Finns were in 
contact, on their southern flank, with the Scythians, who lived west of the 

Finns Carelians Vepses Ingrians Esths Livs Muroma f Merians f Cheremisses 

/^^* r "' 


Zyriand Permiaka 

Voduls Ostu 

1 i ' r 1 






After Kajava, Y., EA, #8, #9, 1922, pp. 353-358. 

Don, and with the Sarmatians, who occupied the plains to the east of it. 
Baltic peoples seem to have touched them on the west, for Baltic words 
are in use among Mordvins, who have never been near the sea. In the time 
of the earliest Greek accounts, Finns seem to have occupied all of the 
country which stretched from the Polessje district of White Russia to the 
central and lower reaches of the Volga. Herodotus located a people called 
Budinoi in the eastern part of this region, presumably in the Volga coun- 
try; west of them he placed the Androphagoi, then the Melancheles 
(Black Mantles), and in the very west, the Neuroi. The name Androphagoi 
or Cannibals, has the same meaning as the Iranian word Mord-Chvar, 
whence are derived Mordva and our own term, Mordvin. The black 
mantle to which Herodotus referred is still a part of the national costume 
of the Volga Finns. 

During the centuries immediately preceding the Christian era, the 
ancestors of the Baltic Finns migrated westward from their original home 

70 The Ijores number roughly 11,000, the Vodes about 700. Exact figures for the 
Evremeiset and Savakot have not been obtained. 

71 The chief source for the following historical resume" is Bunak, V., ZFMA, vol. 30, 
1932, pp. 441-503. 


to the eastern shore of the Baltic, south of the Gulf of Finland, where they 
occupied the country north of the Dtina and the northern half, at least, of 
Kurland, thus taking over most of what is now Latvia, as well as Esthonia. 
After the beginning of the Christian era, some of them crossed the Gulf of 
Finland and settled near Abo and in the Kokemaki and Kyro valleys of the 
present Finland. This country was already inhabited by an Iron Age 
population, of Scandinavian cultural affinity, which the Finns completely 
absorbed. The invaders gradually spread eastward until, about 700 A.D., 
they reached the present Carelia. Thence they went to southern Savo, 
which seems to have been permanently occupied by 1000 A.D. From there 
on the occupation of Finland spread gradually northward until eventually 
the Finns spilled over into Sweden, as related in an earlier section. The 
Finnish penetration of parts of Sweden was only one-half of a reciprocal 
action, however, for even earlier, in the thirteenth century, the Swedes, 
coming by sea, made crusades against the Finns, and many Swedes re- 
mained on the eastern shores of the Gulf of Bothnia and the northern shore 
of the Gulf of Finland. It was at this period that the migration began 
which gave Finland her present Swedish coastal population. Meanwhile 
the ancestors of the Baltic-speaking Letts had moved northward into 
Kurland and Livonia, partly forcing the more southerly Baltic Finns out 
of what is now Latvia, and partly absorbing them. 

Between Leningrad and the Finnish homeland may be seen the rem- 
nants of the early migrant groups, who, when the Slavs first appeared, 
between the sixth and eighth centuries A.D., formed a continuous belt 
of Finnish-speaking peoples. Nearest the Gulf are the Vodcs and Ijores, 
and the Leningrad tribes; on the shores of Lake Onega and the 
headwaters of the Oyat River live the Vepses, who formerly possessed 
a large territory and were a powerful people well into Slavic times. 
To the south and east of the Vepses lived the Merians, now linguis- 
tically extinct, who covered the territory between the Oka and the 
upper Volga. Farther south and east lived the now equally extinct 
Muroma, and then various tribes of Cheremisses, and finally the 
Mordvins. The connecting links between the Vepses and the Chere- 
misses have disappeared, and the groups that have survived have 
suffered great losses of territory. 

The position of the Carelians in this picture is not quite clear; it is 
known, however, that they had settled the shores of the White Sea as early 
as 900 A.D., and were later largely dislodged by Russians. They are lin- 
guistically a branch of the Baltic Finns most closely related to the Estho- 
nians, but it is not known whether they ever were actually in Esthonia, or if 
so, whether they moved northward across the Gulf of Finland with the 
Finns, or around its eastern end. In any case, the Carelians now living in 


Ingria and the Volga country seem to represent a secondary infiltration 
from the present Carelia rather than an early survival. 

Although the departure of the ancestors of the Baltic Finns from their 
Volga homelands took place so early that the movements of the central 
Asiatic nomads did not affect them directly, these incursions were re- 
sponsible for other Finno-Ugrian migrations. In the first century A.D., 
the Huns entered the Volga country and remained along its lower and 
middle course until after having routed the Ostrogoths, when they went 
on to the present Hungary. In the fifth century, after their misadventure 
in France, the Huns returned to the Don Basin and joined their relatives 
the Bolgars, who had come from the region of the Ural and Kuban Rivers 
in southeastern Russia, and had settled between Finns and Ugrians on 
the lower Volga and Kama. There they founded a powerful empire, 
which was to last from the eighth to the fourteenth centuries. Some of 
these Bolgars migrated to the lower Danube country and defeated the 
southern Slavs, settling in what is now Bulgaria. These Bulgarians later 
lost their Uralic speech, and adopted a Slavic language. 

The Bolgars of the Volga ruled or at least influenced a number of Finno- 
Ugrian peoples; the Mordvins, Cheremisses, Votiaks, Syryenians, and 
Magyars. The modern Chuvash of eastern Russia are the linguistic de- 
scendants of the Bolgars, but are thought to be largely Finnish in blood. 
It was at the time of the Bolgar empire or later that the Syryenians moved 
northward, as did the Votiaks, who remained somewhat nearer the center 
of dispersion. Only the Cheremisses and Mordvins still remain in the 
original Finno-Ugrian home territory. 

Under Turkish leaders a large body of Ugri left this region and migrated 
to the southern steppes, whence in the ninth century they moved to 
Hungary, and mixed with the remnants of the Huns and Avars who dwelt 
there. These Ugri became the Magyars, the modern Hungarians, whose 
language is still basically Ugrian, modified by much Turkish influence. 

The closest linguistic relatives of the Magyars are the Voguls and 
Ostiaks, members of primitive hunting and fishing tribes of the Obi country 
in Siberia. By the end of the first millennium A.D., they had moved to the 
northeastern section of European Russia, where they are said to have 
lived with the Samoyeds. The northward movement of Russian colonists 
forced them over into Siberia, and by 1364 they were already entirely 
located on the Asiatic side. Today they are still primitive hunters and 
fishermen, and shamanists in religion. It is believed, however, on philo- 
logical grounds, as well as on historical, that before their migration north- 
ward and eastward they were farmers and herdsmen. 

The Finno-Ugrians today include peoples in every stage of culture 
from hunting and fishing to that of modern civilized states. They are 


held together by a bond of language and by a certain modicum of old 
cultural traits, particularly those concerned with music and poetry, with 
which high and low cultural levels fail to interfere, Although aggressively 
persistent on the peripheries of their radius of migration, they have be- 
come recessive in the center of it, where the great surges of invasion and 
of empire from central Asia, and the subsequent steady and irresistible 
expansion of the eastern Slavs, have reduced their cultural survivors to a 
minimum, while their physical survivors, since they form an important 
element in the composition of modern Russia, are much more numerous. 
In view of their history the Finno-Ugrians are a much more important 
factor in the building of eastern Europe than their present numbers 
would, on the surface, indicate. 


Before studying in detail the physical characters of the Finno-Ugrian- 
speaking peoples apart from the Baltic group and from the Hungarians, 
it will simplify matters considerably to state one fact which the existing 
anthropometric documents make evident: these peoples are all very much alike. 
The Mordvins, the Cheremisses, the Perrnians, the various Ingrian groups 
of the Leningrad region, and the Carelians as well, vary among themselves 
to only a very minor degree. In this respect they differ not only from the 
Baltic Finns and Magyars, but also to a lesser extent from the Ugrian- 
speaking peoples of Siberia. In order to define the basic racial type of 
the present-day central Finnic peoples, it will be easiest to describe that 
one group which has been subjected to the most thorough and most de- 
finitive modern racial study, the Carelians, 72 and afterwards to see how 
the description so obtained applies individually to the other Finnic peoples. 

There are nearly half a million Carelians in Europe; approximately 
half of these live in eastern Finland, and the rest are divided between the 
Carelian Republic of the USSR, which is adjoining, and small ethnic 
islands in the upper Volga country. 78 Zolotarev's adult male sample 
includes 728 from the Carelian Republic, and 277 from the Volga country, 
In both divisions, what he considers dark hair is found among 9 per cent 
of the whole, and the same is true for dark eyes. The Volga group has 
27 per cent of hair designated as quite blond, while the remaining ma- 
jority falls into the light brown and brown category. The Carelians of 
the Republic have, in contrast, 40 per cent of the lightest class. The 
opposite disproportion is true of eye color; 42 per cent of the Volga 
Carelians are called light eyed and 49 per cent mixed; in the Republic 
the figures are 35 per cent and 55 per cent respectively. To begin with, 

72 Zolotarev, D. A., Kareli, SSSR. 

Atlas of Finland, 1925. Zolotarev, TKJP, 1928. 


therefore, the Carelians are typically light or mixed in pigmentation, and 
fully or nearly as blond as most Scandinavians. There is little difference 
in degree of hair and eye pigment between these Finns and Iron Age 
Nordics. The Carelians are prevailingly ash-blond rather than golden, 
and only 4 men out of a thousand show any rufosity. 

The mean stature of the total Carelian group is 165.7 cm., with the 
Volga males slightly taller than those farther north and west. This is a 
moderate or medium height, neither notably tall nor short; it is short by 
Nordic standards, and about average when compared with that of most 
Russians. The bodily proportions as indicated by Zolotarev's data as well 
as by those of other investigators do not show the lateral, heavily built type 
predicted by the study of Finnish influence in Scandinavia; on the contrary, 
the relative sitting height index of 53 is little higher than that expected 
among Nordics. The shoulder and hip diameters are similarly of an in- 
termediate European form. The Carelians are not distinguished by any no- 
table peculiarity in body build, and are more nearly slender than massive. 

The head dimensions, while variable, are smaller than those we have 
found among Scandinavian peoples. The mean length for the total group 
is 187.8 mm., for the Volga sample 186.0 mm.; ten whole millimeters 
shorter, for example, than the Icelandic head length. At the same time 
the breadth, 152.1 mm. for the total group and only 151.7 mm. for the 
Volga Carelians, is less than that of many dolichocephalic Scandinavians. 
The cephalic index, which varies between the extremes of 69 and 90, has 
a mean of 81.1 for the total, 80.9 for the Republic sample, and 81.6 for 
the Volga group. The standard deviation of 3.3 index points for the total 
shows that the Carelians form a reasonably homogeneous group in this re- 
spect. The head form of these basic Finns is therefore sub-brachycephalic, 
or falls into an extremely high mesocephalic category. In size, as well 
as in proportions, the Carelian head stands close to the old Neolithic 
Danubian racial standard. The vault elevation of 127 mm. is high, but 
not extremely high; it is equal to that of Iron Age Nordics in Scandinavia, 
and comparable to that found on the skull among Danubians. 

The forehead breadth of 105.8 mm. is again equal to that of Nordics, 
while the bizygomatic of 139.4 mm. is slightly wider; the total face height 
of 120.8 mm. is between short and medium. The facial index of 86.8 
falls into a moderately broad-faced category. The nose is absolutely quite 
short (50.9 mm.) and of only moderate breadth (35.5 mm.) The resultant 
nasal index, 70.2 in the total group and 69.5 in the Volga sample, lies 
on the borderline between leptorrhiny and mesorrhiny. One of the most 
distinctive measurements in this Carelian group is the interorbital diam- 
eter, with a mean of 34.1 mm., which shows the expected wide-eyed Fin- 
nish form. 


On the whole, the Carelian sample shows nothing in common metrically 
with the large-headed mesocephalic and brachycephalic populations of 
western Norway and Denmark; it may be compared, however, most 
profitably with the Iron Age Nordic type of eastern Scandinavia; in com- 
parison with the latter, the Carelians are short in stature, short in ab- 
solute head length, short in face height and nose height, and slightly 
broad in face breadth. The metrical position of the Carelians among 
living European races is comparable to that of the Danubian type in the 
skeletal series. Both in pigmentation and in basic metrical character it 
shows a certain fundamental relationship to the Iron Age Nordic form. 

The observations tabulated by Zolotarev confirm this general impres- 
sion. The facial outline is called rectangular in 55 per cent of the series, 
and ovoid in 33 per cent; the nasal profile is straight in half the sample, 
concave in 40 per cent, and convex in the remaining tenth. The tip of the 
nose points upward twice as frequently as downward; the lateral profile 
of the forehead is as a rule steep; in only one out of ten instances are the 
two profile lines parallel, as in the characteristic Nordic form. Mongoloid 
features, including an internal eyefold and extreme malar projection, 
are not typical, but are more frequent in the series from the Garelian 
Republic than in that from the Volga country. Only six men out 
of 1008 have the true Mongolian eyefold and these are all in the Repub- 
lic series. 

From these observations as from the measurements, we derive a com- 
posite picture of a moderately variable racial type which is more blond 
than brunet, but prevailingly light mixed in pigment character; square 
or oval faced, with a straight to concave snub-tipped nose, a steep, often 
protruberant forehead, and only moderately projecting malars. A slight 
facial flatness gives a superficial mongoloid impression, but evolved mon- 
goloid features are usually lacking. Throughout there is an incipient 
Nordic suggestion, and in roughly ten per cent of the whole, the Nordic 
head form and facial features, with a longer, elliptical face and parallel 
forehead and nasal profiles, appear. There is undoubtedly a submerged 
Nordic element here, as well as a lesser mongoloid one. 

In view of the general position of the central Finnic type, as exemplified 
by these Carelians, it has seemed most in accordance with the facts to 
leave the designation East Baltic for the larger-bodied, larger-headed, 
and quite different population of the eastern Baltic states, whether Finnic 
or Baltic-speaking, and not to attach it to this clearly differentiated racial 
group which has its geographical center elsewhere. In view of the close 
similarity between this central Finnic type and the Danubian racial en- 
tity suggested -by the early skeletal material, and in view of the fact that 
the earliest identifiable Finnic skeletal remains were mostly of this type, 


it has seemed appropriate, as stated in the concluding section of Chap- 
ter VIII, to name this racial type Neo-Danubian. 

In the early Finnish skeletal remains there was evidence of considerable 
admixture with wide-eyed, broad-faced, meso- and brachycephals from 
the northern forest, and to a lesser extent with what seemed to be Corded 
people or evolved Nordics; these same admixtures are equally apparent 
in the present amalgam, which however remains, for the most part, of 
the same basic racial type to which the earliest agriculturalists to enter 
central Europe overland from the east belonged. It is a mistake to asso- 
ciate the origins of the Finnic people and Finnic speech with a forest 
culture, on the basis of modern associations; the Finns were from the start 
agriculturalists, and have remained such when circumstances have per- 

Let us now study the other Finnic groups inhabiting Russian territory. 
Data on the physical characters of Ingrians are extremely scarce; the 
Ijores, 74 with a mean of 165.6 cm., equal the Carelians in stature, while 
the two Leningrad tribes, the Evremeiset and Savakot, 75 are taller 
(167.1 cm.). The Ijores have a mean cephalic index of 82.6, the Vodes 76 
of 83.2. Thus the deviation of the Ingrians, as shown by this evidence 
from the Carelian standard, is in the direction of brachycephaly. The 
Vepses, who live in a more northerly habitat, are close to the Carelian 
means in these two criteria, with 164.0 cm. for stature, and 81.9 for the 
cephalic index. 77 Observations on a small series of Vepses, 78 however, 
show a majority of brown hair shades, of gray eyes, of broad noses, and 
of oblique eyes, with a weak beard development in many cases, indicating 
a higher Mongoloid content in this group exposed to Lappish and Sam- 
oyed influences, than in most other Finnic samples. 

Of the original Volga Finns, but two tribes, the Mordvins and the 
Cheremisses, retain their ethnic identities, while still living in the center 
of the original Finnish territory. The Mordvins are scattered in single 
villages and groups of villages throughout the middle Volga provinces; 
these settlements, although not continuous, contain collectively a large 
population, officially enumerated by Zolotarev at 1, 167,537. 79 Some 
35,000 more live apart from their own people in the Bashkir and Tatar 
republics, while over 50,000 more have been settled in Siberia and in the 
central Asiatic khanates. Since the Mordvins are excellent farmers and 

74 Zolotarev, D. A., TRIP, 1928, after Prelov, E. I., Alexandrova, A. I., and UP, 
E. F. 

76 Ibid. 

78 Zolotarev, D. A., Kareli, SSSR, after Alexandrova. 

77 Zolotarev, D. A., Kareli, SSSR, after Rozov. 

78 Mainov, V. N., 1877, from a r6sum in AFA, vol. 11, 1879, p. 329. 
w Zolotarev, D. A., TKIP, #15, 1928. 


hardy colonists, they were sent eastward in large numbers by the czars 
to settle newly opened agricultural lands. 

The Cheremisses, who call themselves Mari, number about half a mil- 
lion, and live to the north of the Mordvins in the neighborhood of Kazan. 
Owing to its compactness, their territory has been given the status of an 
autonomous district. They are usually divided into two groups, the Forest 
Cheremisses and the Mountain Cheremisses; the former live in the low- 
lands on the western bank of the Volga, while the latter inhabit a more 
isolated territory to the east, where they preserve many pagan customs. 
Besides following the usual Finnish pursuits of farming and bee culture, 
the Cheremisses, like the Siberian Ugri, are also hunters and stream 

The Mordvins and Cheremisses resemble each other closely in an 
anthropometric sense, and both in turn deviate but little from the stand- 
ard established by our study of the Carelians. 80 The Mordvin stature mean 
is 166.4 cm., that of the Cheremisses 163.7 cm. In bodily proportions the 
Carelian similarity seems complete; in head dimensions the only difference 
is that the Mordvin vault (134 mm.) and that of the Cheremisses (130 mm.) 
may be slightly higher, although these differences may in part be due to 
technical factors. The faces of the Mordvins and Cheremisses are again 
slightly larger than those of the Carelians, with nasion-nienton heights of 
124 and 123 mm., and bizygomatic diameters of 141 and 140 mm. The 
Mordvin nasal index mean, 65.4, is leptorrhine, while that of the Cher- 
emisses, 71.4, is mesorrhine. 

In pigmentation and in soft part morphology, these Volga Finns re- 
semble the Carelians less closely. Bunak, Sergeev, and Mainov find re- 
spectively 33 per cent, 52 per cent, and 60 per cent of light eyes among 
the Mordvins; while there is no specific information regarding the hair 
color of these people, Bunak's statement that 50 per cent of his series 
belongs to a brunet pigment type would indicate that brown was the 
commonest color. Among the Cheremisses, Sommier finds 28 per cent 
of light eyes as against 39 per cent of dark ones; 21 per cent of "light 
blond" hair, and 35 per cent which is dark brown and black. Thus the 
Cheremisses appear to be darker than their more southerly neighbors, and 
both darker than the Carelians. A special series of eastern or mountain 
Cheremisses, measured by Nikolsky, shows clear differences from the major 

80 The chief sources for Mordvins and Cheremisses are : 

Bunak, V., RAJ, vol. 13, 1924, pp. 178-207. 

Sergeev, V. I., PCZA, 1930, pp. 318-319. 

Older works include : 

Maliev, N., r6sum6 in AFA, vol. 12, 1880, p. 392. 

Nikolsky, B., r6sum6 in AFA, vol. 26, 1899, pp. 187-190. 

Sommier, S., APA, vol. 18, 1888, pp. 215-257. 


group with a mean stature of 167.4 cm., a cephalic index of 78.6, and 
60 per cent of blue and gray eyes, and only 32 per cent of black and dark 
brown hair. 

Observations of statistical value which describe these people are scarce. 
However, there seems to be a moderately high incidence of concavity of 
the nasal profile, 18 per cent among Mordvins and 39 per cent among 
Gheremisses; of a median eyefold, which is a sign, as a rule, of a low bony 
orbit 34 per cent among Mordvins and 46 per cent among Cheremisses; 
and 64 per cent of weak beard growth among Mordvins, and 77 per cent 
among Cheremisses. In general, the Cheremisses seem more mongoloid 
than the Mordvins, but on the other hand the isolated Forest Cheremisses 
preserve the least mongoloid type of all, and that closest to a Carelian 
and to a Nordic form. The implication is that while both Mordvins and 
Cheremisses preserve their original Finnic type with considerable fidelity, 
the infiltration of Mongol and Tatar peoples into their country since the 
time that the ancestors of the Carelians and other western tribes departed 
has had some recognizable effect upon them. 

Parallel, in linguistic taxonomy, to the combined Baltic and Volga 
Finnic group is that of the Permians. These are divided into three living 
peoples, the Votiaks, Syryenians (or Zyrians), and Permiaks. All of 
these peoples live north of the Mordvins and Cheremisses, from whose 
general area they are said to have migrated. The most southerly are the 
Votiaks, who, numbering approximately half a million, live on the banks 
of the Kama River, a branch of the Volga, in the southeastern part of 
the former Viatka government. This region has been made into the 
Votiak Autonomous S. S. District by the Soviet authorities. Some 25,000 
other Votiaks live in the Bashkir Republic, 20,000 in the Samarsk govern- 
ment, 1700 in Siberia, and others still in the Tatar Republic. In general, 
the modern destiny of the Votiaks has been to a certain extent associated 
with that of Turkish-speaking peoples. In their own language they call 
themselves Udmurt, and this language contains many loan words from 
Chuvash and Tatar speech. They have, however, failed to become Moslem; 
their religion, at the time of the Russian revolution, was officially Orthodox 
Christianity, which served as a cloak for the retention of much of the 
original Finnish heathendom. 

Metrically the Votiaks resemble the Cheremisses very closely. 81 A 
cephalic index mean of 82 is slightly higher, and reflects a slightly smaller 
head length. The mean stature is 162 to 163 cm. The pigment characters 

81 Chief sources on the Votiaks are : 

Chomiakov, M. N., TKU, vol. 43, #3, pp. 1-294. Rsum6 in ZBFA, vol. 17, 1912. 

Maliev, N., TKU, vol. 4, #2, 1874, pp. 1-17. R6sum6 in AFA, vol. 9, 1876, p. 227. 

Khonuakov, 1911, after Zolotarev, TKIP, 1928. 

Teploukhov, S. A., after Zolotarev, TKIP, 1928. 


of the iris are similar to those of the Mordvins and Cheremisses, since 
between 30 per cent and 35 per cent of eyes are called brown, and the 
rest divided between blue, gray, and mixed colors. In head hair color 
however, a difference may be seen, for Maliev's series shows that but 
2 per cent are black, 32 per cent dark brown, 29 per cent brown, 15 per 
cent light brown, and 7 per cent flaxen. Of the rest, 11 per cent are listed 
as reddish-brown. Chomiakov confirms this high incidence of rufosity, 
with 6 per cent of red hair color. Among Maliev's subjects only 1 5 per cent 
had black -or brown beards; of the others 47 per cent were listed as red. 
These Votiaks, then, are not as blond as the Garelians, but blondism is 
frequent and characteristic; rufosity, notably absent from both the Car- 
elian group and from the Iron Age Nordic race, and not important among 
the two tribes of Volga Finns, becomes a major factor among Permians. 

The Votiaks are usually deficient in body hair, and the beard is fre- 
quently sparse, although in individual cases very heavy beards and very 
abundant body hair are found. The Ainu-like pilosity of many Russian 
peasants is commoner among Slavs than among Finnic speakers, but is 
exceptional in both groups. Neither, by and large, are as hairy as most 
western European brachycephals. The hair form is predominantly 
straight, only exceptionally wavy or curly. Forty per cent of Votiaks are 
listed as long- or oval-faced; the remainder as round-, broad-, or flattish- 
faced. The nose is straight in 60 per cent of individuals, convex in but 
12 per cent. Maliev states that 37 per cent are "solid" in bodily build, 
only 6 per cent linear or thin. All in all the Votiaks are typical Finns, 
slightly shorter and rounder headed than Carelians or Mordvins, oddly 
rufous, and not noticeably more mongoloid than their southerly neighbors. 

North of the Votiaks live two allied tribes, the Syryenians and Permiaks, 
both of whom call themselves Komi. These two peoples are generally 
lumped into a single category, especially since they speak mutually in- 
telligible languages and occupy contiguous territories. The Syryenians 
occupy the wide expanse stretching from 58 N. Latitude to the Arctic 
Ocean, and from the Ural Mountains on the east to the Pinega River, a 
tributary of the Dvina, on the west. There are also a few Syryenians who 
live on the Siberian side of the Urals. The chief town of the Syryenians is 
Ishma, on the Pechora River. The Permians live more on the eastern side 
of the upper Kama River. Population statistics regarding these peoples are 
very variable. Zolotarev gives 186,000 as the total for Syryenians living 
in Russia, and 9566 for Siberia. Jochelson estimates the Syryenians at 
260,000, and the Atlas of Finland at 364,000. Zolotarev finds 130,000 
Permiaks in the Komipermiaktsk and Berhuekamsk districts of Uralsk 

Below 65 N. Latitude the Syryenians and Permiaks farm, and are noted 


for their skill and perseverance in obtaining crops at such high latitudes. 
Beyond the line at which agriculture becomes impossible, the Syryenians 
breed reindeer and live a less settled existence. They are noted for their 
ability at trading and their general financial sharpness. In religion they 
are said to adhere strictly to Orthodox tenets and to have forsworn the 
pagan practices which linger on among the Votiaks. 

In stature, in body build, and in head dimensions and proportions 
both the Permiaks and Syryenians seem to be identical with their rela- 
tives the Votiaks; 82 a difference between these peoples and the Volga 
Finns, however, may exist in nose form, for the nasal index mean of the 
Permiaks is 64. 9, 82 of the Syryenians 65. 7. 82 Only 1 1 per cent of Syryenians, 
and 14 per cent of Permiaks, are said to have dark eyes; thus these north- 
ern Permians are perhaps both lighter eyed and more leptorrhine than 
most of the Volga Finnic group. One sub-group of Syryenians, living in 
the Ust-Sylosk district, seems to have mixed with Samoyeds or other non- 
Finnic peoples, for the cephalic index is 83.3, as contrasted with the usual 
mean of 81 for other Syryenians, and the ratio of dark eyes is twice that 
for the others. 

In hair color, for the Syryenians as a whole, we find at last a series of 
observations based on the Fischer chart, and taken on a series of 400 
individuals. 83 The authors divide the scale into three categories; dark 
(Fischer #4-8), golden (#9-15), and ashen (#16-26). The percentages 
are 53.1 per cent, 9.6 per cent, and 37.4 per cent. It is to be noticed that 
no individuals were listed as black, or as red. A medium or dark brown 
is the most numerous shade, with ash-blond next commonest. One is led, 
in view of this, to suspect that the high degree of rufosity reported among 
Votiaks may be partly of technical origin. 

Summarizing the data on these Permian speakers, we may state that 
they seem to resemble the Carelian norm more completely in head and 
face form, and in pigmentation, than do the Cheremisses or Mordvins. 
It seems likely that those Finns and Permians who dispersed from their 
homes during the early centuries of the present era in a northward as well 
as in a westward direction carried with them the older Finnish features, 
while those who remained in their Volga home were to a greater extent 
affected by Tatar and other influences. On the whole, however, the 

82 Sources on these peoples are : 

Alexandrova, A., Nurk, L., and U1 J , E., PCZA, 1930, pp. 287-288. 

Ivanovsky, A. L., AFA, vol. 48, 1925, pp. 1-12. 

Maliev, N., TKU, vol. 16, #4, 1887. 

Naiimov (after Zolotarev, TKJP, 1928). 

Sevastianov (after Zolotarev, TKIP, 1928). 

Vishnevsky, B., Anthropologicheskifa Dannifa o Naselenil Permskaga 

88 Alexandrova, Nurk, and UP, op. cit. 


generalization that the entire body of Finnic and Permian speakers, 
apart from the Baltic groups which remain to be studied, are closely unified 
in race has been shown to be accurate. 

Before concluding this survey of eastern Finno-Ugrian peoples, one 
further group requires examination, that of the Ostiaks and Voguls, the 
Ob-Ugrians, the primitive hunters and fishers of western Siberia, who 
are the closest linguistic relatives of the Magyars. The Ostiaks have been 
reduced to less than 20,000 individuals, while the Voguls number be- 
tween 5000 and 7000. Some of the Ostiaks have been thoroughly Russian- 
ized, while others have mixed with Samoyeds, and have taken over 
reindeer breeding. At the time of the Russian expansion eastward into 
Siberia, the Ostiaks were the first to feel the pressure, and hence the 
southern part of their territory was taken from them, and they were re- 
duced to more primitive circumstances than before. 

One must expect the modern Ostiaks and Voguls to show the effects 
of centuries of reduced conditions of living, and this is, indeed, manifested 
in their reduction in stature; various means place them at levels between 
154 and 160 cm., but the largest series of both groups fall in the 158-159 
cm. category. The bodily form of both is in most cases slight and lean. 84 
Sommier, with a series of 106 male Ostiaks, finds 50 per cent to have 
brown eyes, and the rest mixed and light; the Voguls are apparently 
somewhat lighter eyed. 86 About 25 per cent of Ostiaks have light brown 
or blond hair, and the Voguls are again slightly fairer. 86 In both groups 
and in all series, black hair is very much in the minority. Ten per 
cent of it among Ostiaks may well indicate Samoyed admixture. On 
the whole, these Siberian forest Ugrians are the darkest of the Finno- 
Ugrian-speaking peoples, aside from the Lapps, whom we have already 

The head form of the two Siberian tribes is the same as that of 
the Volga Finns and Permians, the dimensions somewhat smaller. The 
faces seem to be shorter, and the noses are definitely mesorrhine. Pho- 
tographic evidence makes it certain that both the Voguls and Ostiaks 
have absorbed a perceptible amount of mongoloid blood, which man- 
ifests itself especially in facial features. They are still, however, bas- 

84 Chief sources on Ostiaks and Voguls are : 

Maliev, N. M., RAJ, vol. 5, 1901, pp. 73-81. 

Rudenko, S. L, BMSA, vol. 6, ser. 5, 1914, pp. 123-143. 

Sommier, S., APA, vol. 17, 1887, pp. 71-222. t , 

86 Rudenko finds 87 per cent of brown eyes among 53 Ostiaks, 76 per cent among 
75 Voguls 

86 Rudenko, again, has 96 per cent brown hair for Ostiaks, 81 per cent for Voguls. I 
have taken Sommier's figures in preference to Rudenko's because Sommier's series are 
larger and his hair and eye color classifications more detailed, permitting judgment 
and recombination. 


ically similar in most characters to their relatives on the other side of the 
Urals. 87 


For the sake of continuity, let us return to the beginning of section eight, 
in which we expressed the intention of studying the racial composition 
of the Baltic Finns. We have seen, in the meanwhile, that the basic Finnic 
racial type, to which belong the Volga Finns and their relatives from 
Carelia to the Obi River, is a modern counterpart of the prehistoric 
Danubian race, with leanings in both a Nordic and a Ladogan direction. 
This type has been, therefore, named Neo-Danubian. It was this Neo- 
Danubian racial type that the ancestors of the Balt